JP5299159B2 - Imaging apparatus and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an imaging apparatus in which all defective pixels to be corrected can be corrected, regardless of the limited number of registration in a storage part, and to provide a program for defective pixel correction. <P>SOLUTION: The imaging apparatus includes an image generation means (28) for generating image data, based on a plurality of electrical signals output from an imaging device (20) having a plurality of pixels, wherein the image generation means has a defective pixel correction part 42 which corrects defective pixel of image generation data by a defective pixel correction operation. The image generation means sets the number of correction pixels to be corrected, based on the defective pixel level, associated with criteria of change in influences of the defective pixels; and when the number of correction pixels exceeds the limited number of registration which is the number of pieces of defective pixel data that can be registered in the storage part 43, the image generation means repeats the defective pixel correction operation in the defective pixel correction part 42, while changing the registered content of the defective pixel data in the storage section 43 so as to correct the defective pixel of the number of correction pixels, without making the pixel data to be corrected duplicate. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to an imaging device and a program, and more particularly to an imaging device such as a digital still camera having a defective pixel correction function and a program for correcting the defective pixel.

  In an image sensor, such as a CCD, used in a digital still camera, not all pixels are formed with completely the same performance in the semiconductor manufacturing process, and defective pixels are also formed. However, even if there are a predetermined number or less of defective pixels due to the product yield, it is generally used as a good CCD for a product such as a digital still camera. For this reason, in the product, the ordinate and abscissa (address) on the CCD of the defective pixel (hereinafter referred to as defective pixel data) are stored in advance in the storage unit of the defective pixel correction unit provided in the product, and the defect is detected at the time of shooting. In the pixel correction unit, the pixel data corresponding to the address of the defective pixel stored in the storage unit in the imaging data is interpolated with the pixel data of the same adjacent color filter, and then the defective pixel correction such as still image processing It is considered to carry out the processing.

  In addition to the above-described defective pixels, there are also defective pixels that have poor dark current characteristics (temperature scratches) in the CCD. Since the defective pixel due to the dark current is a dynamic defect factor, when the defective pixel correction process is performed as described above, the defective pixel (and its pixel) to be corrected in the CCD is determined by the shooting parameters such as the exposure time. It is considered to change the number) and correct defective pixels accordingly (see, for example, Patent Document 1).

  By the way, in such an imaging apparatus, it is required to reduce the size of the entire apparatus as much as possible, and a mechanism for various functions is accommodated therein, so that defective pixel correction processing is performed. Therefore, the number of defective pixel data (defective pixel number) that can be stored in the storage unit is limited. .

  For this reason, if the number of defective pixels to be corrected in the imaging device exceeds the number of defective pixels (registration limit number) that can be registered in the storage unit of the defective pixel correcting unit, all defective pixels to be corrected are It becomes impossible to correct.

  The present invention has been made in view of the above problems, and an imaging apparatus capable of correcting all defective pixels to be corrected regardless of the registration limit number in the storage unit and the defective pixel correction thereof It aims to provide a program for.

In order to solve the above-described problem, the invention according to claim 1 includes image generation means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels, and the image generation means comprises: When image generation data in the generation process of the image data is input, a defective pixel correction operation for correcting pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data, An imaging apparatus having a defective pixel correction unit that performs defective pixel correction of the image generation data, wherein the image generation means is a correction target based on a defective pixel level associated with a determination criterion for a change in the influence of a defective pixel. When the number of correction pixels to be set is set and the number of correction pixels exceeds the registration limit number that is the number of the defective pixel data that can be registered in the storage unit, the correction target number is set. In order to perform the correction pixel number of the defective pixel correction without overlapping raw data, with repeating the defective pixel correction operation in the defective pixel correction unit while changing the registration contents of the defective pixel data in the storage unit, The image generation data is divided into a plurality of divided image data by equally dividing the number of corrected pixels by the value obtained by dividing the correction pixel number by the registration limit number, and dividing each divided image data. In addition, the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registration contents of the defective pixel data in the storage unit .

The invention according to claim 2 includes image generation means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels, wherein the image generation means is in the process of generating the image data. When image generation data is input, defective pixels of the image generation data are obtained by a defective pixel correction operation that corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. An imaging apparatus having a defective pixel correction unit that performs correction, wherein the image generation unit sets the number of correction pixels to be corrected based on a defective pixel level associated with a criterion for determining a change in the influence of a defective pixel. When the number of corrected pixels exceeds the registration limit number that is the number of defective pixel data that can be registered in the storage unit, the pixel data to be corrected may be duplicated. In order to perform defective pixel correction of the number of corrected pixels, the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registered content of the defective pixel data in the storage unit, and the image generation data Then, the defective pixel data of the registration limit number is registered in the storage unit and the defective pixel correction unit performs the defective pixel correction operation, and then the image generation data is stored in the registration limit of the correction pixel number. The difference obtained by subtracting the number is divided by the registration limit number and divided into a plurality of divided image data by equally dividing the value obtained by rounding up the fractional number of the quotient, and defective pixels in the first defective pixel correction operation Correction refers to the previous correction in the defective pixel correction unit while changing the registration contents of the defective pixel data in the storage unit for each divided image data so that the pixel data to be corrected does not overlap. And repeating the defective pixel correction operations.

The invention according to claim 3 includes image generation means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels, wherein the image generation means is in the process of generating the image data. When image generation data is input, defective pixels of the image generation data are obtained by a defective pixel correction operation that corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. An imaging apparatus having a defective pixel correction unit that performs correction, wherein the image generation unit sets the number of correction pixels to be corrected based on a defective pixel level associated with a criterion for determining a change in the influence of a defective pixel. , this the correction pixel number, if it exceeds registration limit number is the number of the defective pixel data can be registered in the storage unit, to overlap pixel data to be corrected In order to perform defective pixel correction of the number of corrected pixels, the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registered content of the defective pixel data in the storage unit, and the image generation data Then, in order from the highest defective pixel level, the defective pixel data of the registration limit number is registered in the storage unit and the defective pixel correction unit performs the defective pixel correction operation. The difference between the correction pixel number and the registration limit number is divided into a plurality of divided image data by equally dividing the difference obtained by dividing the difference obtained by dividing the registration limit number into a plurality of divided image data . the registration contents of the defective pixel data in the storage unit so that the pixel data to be corrected does not overlap each said divided image data from the defective pixel correction in the defective pixel correction operation Additional and repeating the defective pixel correction operation in the defective pixel correction unit while.

The invention according to claim 4 is the imaging apparatus according to any one of claims 1 to 3 , further comprising a memory that can be appropriately stored and read by the image generation unit, and the defect The pixel data determines the defective pixel level based on an electrical signal output from the imaging element, and the imaging of the pixel in which the defective pixel level exceeds a correction determination value associated with a setting condition in the determination criterion The ordinate and abscissa on the element are generated in association with the corresponding setting condition and the defective pixel level and stored in the memory, and the image generation means reads out the defective pixel data that conforms to the setting condition from the memory. And registering in the storage unit .

The invention according to claim 5 is the imaging device according to any one of claims 1 to claim 4, wherein the defective pixel data, on the image sensor corresponding to the divided image data divided equally The defective pixel level is determined on the basis of the electrical signal output from the imaging device for each of the sections, and the defective pixel level exceeds the correction determination value associated with the setting condition in the determination criterion. The vertical and horizontal coordinates on the image sensor are generated in association with the corresponding setting condition and the defective pixel level, and are stored in the memory.

The invention according to claim 6 is the imaging device according to any one of claims 1 to claim 5, wherein the criterion is the exposure time, the setting condition is a value set as the exposure time It is characterized by being.

The invention according to claim 7 is the imaging device according to any one of claims 1 to claim 5, wherein the criterion is ISO sensitivity, the setting condition is set number as ISO sensitivity It is characterized by being.

The invention according to claim 8 includes image generation means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels, and the image generation means is in the generation process of the image data. When image generation data is input, defective pixels of the image generation data are obtained by a defective pixel correction operation that corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. A program for causing a control unit of an imaging apparatus having a defective pixel correction unit to perform correction to realize a function of correcting defective pixels by the defective pixel correction unit. A function for setting the number of correction pixels to be corrected based on the defective pixel level associated with the determination criterion, and the number of correction pixels registered in the storage unit When the registration limit number, which is the number of defective pixel data, is exceeded, the defective pixel data in the storage unit is corrected in order to perform defective pixel correction of the corrected pixel number without overlapping pixel data to be corrected. While repeating the defective pixel correction operation in the defective pixel correction unit while changing the registration content, the image generation data, less than the decimal of the quotient obtained by dividing the correction pixel number by the registration limit number Dividing into a plurality of divided image data by equally dividing by the rounded-up value, and changing the registered content of the defective pixel data in the storage unit for each divided image data, the defective pixel correction in the defective pixel correction unit And a function of repeating the operation .

The invention according to claim 9 includes image generation means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels, and the image generation means is in the generation process of the image data. When image generation data is input, defective pixels of the image generation data are obtained by a defective pixel correction operation that corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. the control unit of the imaging apparatus having the defective pixel correction unit that performs correction, a program for realizing a function of correcting defective pixels by the defective pixel correction unit, the image generating means, a change in the influence of the defective pixel A function for setting the number of correction pixels to be corrected based on the defective pixel level associated with the determination criterion, and the number of correction pixels registered in the storage unit When the registration limit number, which is the number of defective pixel data, is exceeded, the defective pixel data in the storage unit is corrected in order to perform defective pixel correction of the corrected pixel number without overlapping pixel data to be corrected. The defective pixel correction operation in the defective pixel correction unit is repeated while changing registration contents, and the defective pixel data of the registration limit number is registered in the storage unit for the image generation data. A quotient obtained by causing the pixel correction unit to perform the defective pixel correction operation and then dividing the image generation data by a difference obtained by subtracting the registration limit number from the correction pixel number by the registration limit number. Are divided into a plurality of divided image data by dividing them by a value rounded up to the nearest decimal number, and the pixel data to be corrected do not overlap with the defective pixel correction in the first defective pixel correction operation. Characterized in that to realize a function of the Ru was repeated defective pixel correction operation in the defective pixel correction unit while changing the registration contents of the defective pixel data in the storage unit for each of the divided image data into .

The invention according to claim 10 is the program according to claim 8 or 9 , wherein the image generation means determines the defective pixel level based on an electrical signal output from the image sensor, The memory is generated by associating the corresponding setting condition and the defective pixel level with the vertical and horizontal coordinates on the image sensor of the pixel whose defective pixel level exceeds the correction determination value associated with the setting condition in the determination criterion. among the defective pixel data stored in the function to register a compatible the defective pixel data in the storage unit is read out from said memory to said setting condition, characterized in that to realize.

An invention according to an eleventh aspect is the program according to any one of the eighth to tenth aspects , wherein the defective pixel data is divided on the imaging element corresponding to the divided image data equally divided. For each section, the defective pixel level is determined based on the electrical signal output from the imaging device, and the defective pixel level of the pixel whose correction pixel value exceeds the correction determination value associated with the setting condition in the determination criterion A function of associating the corresponding setting condition and the defective pixel level with the vertical and horizontal coordinates on the image sensor and storing them in the memory is realized.

  According to the imaging apparatus according to the present invention, even when the set number of correction pixels to be corrected exceeds the registration limit number, by repeating the defective pixel correction operation in the defective pixel correction unit, the number exceeding the number Are corrected by the defective pixel correction operation in the defective pixel correction unit for the second time or later (2 times depending on the setting conditions), so the relationship between the number of corrected pixels that need to be corrected and the registration limit number of the storage unit Regardless, all defective pixels of the image sensor can be corrected.

  In addition, when repeating the defective pixel correction operation in the defective pixel correction unit, the registration content of the defective pixel data in the storage unit is set so that defective pixel correction of the number of corrected pixels is performed without overlapping pixel data to be corrected. Since the change is made, it is possible to correct only the set number of correction pixels, in other words, to correct appropriately without over- and under-correction, and to prevent the image quality from being deteriorated due to insufficient correction or over-correction of defective pixels.

  In addition to the above-described configuration, the image generation unit further includes a memory that can be appropriately stored and read, and the defective pixel data is determined based on an electric signal output from the image sensor. The defective pixel level is generated by associating the corresponding setting condition and the defective pixel level with the vertical and horizontal coordinates on the image sensor of the pixel whose corrected pixel value exceeds the correction determination value associated with the setting condition in the determination criterion. Stored in the memory, and the image generation means reads out the defective pixel data conforming to the set condition from the memory and registers it in the storage unit, the defective pixel correction unit performs a defective pixel correction operation. When repeating, defective pixel data that matches the setting condition read from the memory is registered in the storage unit. Defective pixel correction can be performed.

  In addition to the configuration described above, the image generation means can be obtained by dividing the image generation data by dividing the correction pixel number by the registration limit number when the set correction pixel number exceeds the registration limit number. The fractional pixel is divided into a plurality of divided image data by dividing the value by rounding up the decimal number, and the defective pixel correction unit changes the registered content of the defective pixel data in the storage unit for each of the divided image data When the defective pixel correction operation in step S3 is repeated, since the processing time required for defective pixel correction in the defective pixel correction unit is substantially proportional to the amount of input data, when correcting all defective pixels of the image sensor Regardless of the number of defective pixel correction operations in the defective pixel correction unit, the processing time for the image generation data before being divided can be substantially equal to the case of performing defective pixel correction processing in the defective pixel correction unit. That.

  In addition to the above-described configuration, the image generating means stores the defective pixel data of the registration limit number with respect to the image generation data when the set correction pixel number exceeds the registration limit number. The defective pixel correction unit performs the defective pixel correction operation, and then the image generation data is obtained by dividing the difference obtained by subtracting the registration limit number from the correction pixel number by the registration limit number. Each of the obtained quotients is divided into a plurality of divided image data by dividing the value by rounding up the decimal number, and each of the pixel data to be corrected does not overlap with the defective pixel correction in the first defective pixel correction operation. When the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registration content of the defective pixel data in the storage unit for each divided image data, first, the entire area of the imaging element is applied. By correcting the defective pixels of the recording limit number, it is possible to correct the defective pixels more appropriately, and by performing defective pixel correction for the remaining number of corrected pixels on each divided image data, It is possible to suppress an increase in processing time required for correcting defective pixels for the number of correction pixels.

  In addition to the above-described configuration, when the set number of corrected pixels exceeds the registration limit number, the image generation unit sets the registration limit number in order from the highest defective pixel level with respect to the image generation data. The defective pixel data is registered in the storage unit, the defective pixel correction unit performs the defective pixel correction operation, and then the image generation data is obtained by subtracting the registration limit number from the correction pixel number. Dividing by the value obtained by dividing the quotient obtained by dividing by the registration limit number into equal parts and dividing it into a plurality of divided image data. The defective pixel correction in the first defective pixel correction operation is a correction target. The defective pixel correction operation in the defective pixel correction unit is repeated while changing the registration content of the defective pixel data in the storage unit for each divided image data so that the pixel data to be overlapped does not overlap. First, the defective pixels can be corrected more appropriately by correcting the defective pixels with respect to the entire area of the image sensor by the registration limit number in order from the highest defective pixel level, and the remaining corrected pixels thereafter. By performing defect pixel correction for several divided image data on each divided image data, an increase in processing time required for defect pixel correction for the remaining number of corrected pixels can be suppressed.

  In addition to the above-described configuration, the defective pixel level is determined based on an electrical signal output from the image sensor for each section on the image sensor corresponding to the divided image data that has been equally divided. And correlating the corresponding setting condition and the defective pixel level with the vertical and horizontal coordinates on the image sensor of the pixel whose defective pixel level exceeds the correction determination value associated with the setting condition in the determination criterion. If it is generated and stored in the memory, the defective pixel data in each divided image data according to the setting condition can be appropriately registered in the storage unit of the defective pixel correction unit, and the defective pixel correction process Therefore, it is possible to efficiently store each defective pixel data for the memory in the memory.

  In addition to the above-described configuration, assuming that the determination criterion is an exposure time and the setting condition is a numerical value set as the exposure time, a defective pixel can be appropriately corrected according to the exposure time at the time of shooting. it can.

  In addition to the above-described configuration, if the determination criterion is ISO sensitivity and the setting condition is a numerical value set as ISO sensitivity, defective pixels can be corrected appropriately according to the ISO sensitivity at the time of shooting. it can.

BRIEF DESCRIPTION OF THE DRAWINGS It is explanatory drawing which shows the digital still camera (henceforth "digital camera 1") as an example of the imaging device which concerns on Example 1 of this invention, (a) is a front view, (b) is a top view. (C) is a rear view. It is a block diagram which shows the outline | summary of the system configuration | structure of the digital camera 1 of FIG. 2 is an explanatory diagram schematically showing an RGB filter used in the digital camera 1. FIG. Each priority number of data of correction stored RAM44 in the pixel group P _N defective pixel address (vertical and horizontal coordinates on CCD 20) (the number of data of the defective pixel) or the like is an explanatory diagram showing a table. 3 is a flowchart showing details of defective pixel correction processing of Example 1 executed by a control unit 28; It is a flowchart which shows the processing content of step S4 in the flowchart of FIG. FIG. 5 is an explanatory view similar to FIG. 4 showing, in a table form, the number of defective pixel addresses (vertical and horizontal coordinates on the CCD 20) stored in the RAM 44 in each priority correction pixel group _NM, and the like. It is explanatory drawing for demonstrating a mode that the acquired image (image data) was equally divided into S, (a) shows a mode that it divided into 4, (b) has shown a mode that it divided into 2. 10 is a flowchart showing details of defective pixel correction processing according to a second embodiment that is executed by a control unit. It is a flowchart which shows the processing content of step S24 in the flowchart of FIG. The number of data (number of defective pixel data) of defective pixel addresses (vertical and horizontal coordinates on the CCD 20) stored in the RAM 44 in each priority correction pixel group P _N ′ or each priority correction pixel group P _NM ′ is shown in a table. It is explanatory drawing similar to FIG. 4 and FIG. 10 is a flowchart showing details of defective pixel correction processing of Example 3 executed by a control unit 28. It is a flowchart which shows the processing content of step S56 in the flowchart of FIG.

  DESCRIPTION OF EMBODIMENTS Embodiments of an imaging apparatus according to the present invention and a program for defective pixel correction processing executed by the imaging apparatus will be described below with reference to the drawings.

FIG. 1 is an explanatory view showing a digital still camera (hereinafter referred to as “digital camera 1”) as an example of an imaging apparatus according to Embodiment 1 of the present invention, (a) is a front view, and (b). Is a top view, and (c) is a back view. FIG. 2 is a block diagram showing an outline of the system configuration of the digital camera 1 of FIG. FIG. 3 is an explanatory diagram schematically showing an RGB filter used in the digital camera 1. 4, each priority correction number data of the pixel group P _N is stored in the RAM44 in a defective pixel address (vertical and horizontal coordinates on CCD 20) (the number of data of the defective pixel) or the like is an explanatory diagram showing a table.
(Appearance structure of digital camera)
In the digital camera 1 according to the first embodiment, as shown in FIG. 1, a release button (shutter button) 2, a power button 3, and a shooting / playback switching dial 4 are provided on the upper surface, and shooting is performed on the front (front) side. A lens barrel unit 6 having a lens system 5, a strobe light emitting unit (flash) 7, and an optical viewfinder 8 are provided.

In the digital camera 1, on the rear side, a liquid crystal monitor (LCD) 9, an eyepiece 8a of the optical viewfinder 8, a wide-angle zoom (W) switch 10, a telephoto zoom (T) switch 11, a menu (MENU) A button 12, a confirmation button (OK button) 13, and the like are provided. Inside the side surface of the digital camera 1 is provided a memory card storage unit 15 for storing a memory card 14 (see FIG. 2) for storing captured image data. Note that the appearance of the imaging apparatus according to the present invention is not necessarily limited to the first embodiment, and may have another appearance.
(Digital camera system configuration)
As shown in FIG. 2, the digital camera 1 has a CCD 20 as a solid-state imaging device on which a subject image incident through the photographing lens system 5 of the lens barrel unit 6 forms an image on a light receiving surface, and an electric signal ( Analog front-end unit 21 (hereinafter referred to as “AFE unit 21”) that processes analog RGB image signals) into digital signals, signal processing unit 22 that processes digital signals output from AFE unit 21, and temporarily stores data SDRAM 23 that stores control programs, a motor driver 25 that drives the lens barrel unit 6, and a RAM 44 that can store and read data appropriately by a control unit 28 described later. .

  The taking lens system 5 includes a zoom lens, a focus lens, and the like. The lens barrel unit 6 includes a photographic lens system 5, an aperture unit 26, and a mechanical shutter unit 27. The drive units of the photographic lens system 5, the aperture unit 26, and the mechanical shutter unit 27 are driven by a motor driver 25. . The motor driver 25 is driven and controlled by a drive signal from a control unit (CPU) 28 of the signal processing unit 22.

  The CCD 20 is a solid-state image sensor that converts an imaged subject image into an electrical signal (image data) and outputs it. The CCD 20 has an entire light receiving surface divided into a grid-like area called a pixel (hereinafter also referred to as a pixel 20a (see FIG. 3)), and is an image composed of a set of pixel data that is digital data. Data is output as an electrical signal. In the CCD 20, color filters (RGB, CYM, etc.) are provided in each divided area (pixel) so as to form a Bayer array. In the first embodiment, as shown in FIG. 3, the RGB primary colors of the Bayer array are provided. A filter (hereinafter referred to as “RGB filter”) is arranged. Therefore, the CCD 20 outputs an electrical signal (analog RGB image signal) corresponding to the three primary colors of RGB from each pixel 20a. The CCD 20 can also be configured with a CMOS image sensor or the like as a solid-state image sensor.

  The AFE unit 21 is sampled by a TG (timing signal generating unit) 30 that drives the CCD 20, a CDS (correlated double sampling unit) 31 that samples an electrical signal (analog RGB image signal) output from the CCD 20, and the CDS 31. An AGC (analog gain control unit) 32 that adjusts the gain of the signal, and an A / D conversion unit 33 that converts the signal gain-adjusted by the AGC 32 into a digital signal (hereinafter referred to as “RAW-RGB data”).

  The signal processing unit 22 includes a CCD interface (hereinafter referred to as “CCD I / F”) 34, a memory controller 35, a YUV conversion unit 36, a resizing processing unit 37, a display output control unit 38, and a data compression unit 39. , A media interface (hereinafter referred to as “media I / F”) 40, a defective pixel correction unit 42, and a control unit (CPU) 28.

  The CCD I / F 34 outputs a screen horizontal synchronization signal (HD) and a screen vertical synchronization signal (VD) to the TG 30 of the AFE unit 21, and from the A / D conversion unit 33 of the AFE unit 21 according to these synchronization signals. The output RAW-RGB data is captured. The memory controller 35 controls the SDRAM 23. The YUV converter 36 converts the RAW-RGB data captured by the CCD I / F 34 into image data in YUV format that can be displayed and recorded. The resizing processing unit 37 changes the image size according to the size of image data to be displayed or recorded. The display output control unit 38 controls display output of image data. The data compression unit 39 converts the data so that the image data is recorded in the JPEG format or the like. The media I / F 40 writes image data to the memory card 14 or reads image data written to the memory card 14. The defective pixel correction unit 42 corrects defective pixels in the process in which the image data acquired by the CCD 20 propagates through the signal processing unit 22. The correction of the defective pixel will be described in detail later. The control unit 28 performs system control and the like of the entire digital camera 1 based on a control program stored in the ROM 24 in accordance with operation input information from the operation unit 41.

  The operation unit 41 includes a release button 2, a power button 3, a shooting / playback switching dial 4, a wide-angle zoom switch 10, a telephoto zoom switch 11, a menu button 12, and a confirmation button 13 provided on the external surface of the digital camera 1. Etc. (see FIG. 1), and a predetermined operation instruction signal corresponding to the operation of the photographer is input to the control unit 28.

The SDRAM 23 stores RAW-RGB data captured by the CCD I / F 34 and also stores YUV data (YUV format image data) converted by the YUV conversion unit 36. In addition, the data compression unit 39 The image data such as JPEG formation that has been subjected to compression processing is saved. The YUV data is the luminance data (Y), the color difference (U) that is the difference between the luminance data and the blue (B) component data, and the color difference (V) that is the difference between the luminance data and the red (R) component data. ) Information to express color.
(Digital camera monitoring and still image shooting)
Next, the monitoring operation and still image shooting operation of the digital camera 1 will be described. In the still image shooting mode, the digital camera 1 performs a still image shooting operation while executing a monitoring operation as described below.

  First, when the photographer turns on the power button 3 and sets the photographing / playback switching dial 4 to the photographing mode, the digital camera 1 is activated in the recording mode. When the control unit 28 detects that the power button 3 is turned on and the photographing / playback switching dial 4 is set to the photographing mode, the control unit 28 outputs a control signal to the motor driver 25 to allow the lens barrel unit 6 to be photographed. And the CCD 20, the AFE unit 21, the signal processing unit 22, the SDRAM 23, the ROM 24, the liquid crystal monitor 9 and the like are activated.

  Then, the photographing lens system 5 of the lens barrel unit 6 is directed toward the subject, so that the subject image incident through the photographing lens system 5 is formed on the light receiving surface of each pixel of the CCD 20. An electrical signal (analog RGB image signal) corresponding to the subject image output from the CCD 20 is input to the A / D converter 33 via the CDS 31 and the AGC 32, and 12 bits (bits) are output from the A / D converter 33. To RAW-RGB data.

  This RAW-RGB data is taken into the CCD I / F 34 of the signal processing unit 22 and stored in the SDRAM 23 via the memory controller 35. The RAW-RGB data read from the SDRAM 23 is converted into YUV data (YUV signal) by the YUV conversion unit 36 and then stored as YUV data in the SDRAM 23 via the memory controller 35.

  The YUV data read from the SDRAM 23 via the memory controller 35 is sent to the liquid crystal monitor (LCD) 9 via the display output control unit 38, and the photographed image is displayed on the liquid crystal monitor (LCD) 9. At the time of monitoring in which a captured image is displayed on the liquid crystal monitor (LCD) 9, one frame is read out in a time of 1/30 seconds by the pixel number thinning process by the CCD I / F 34. For this reason, the AFE unit 21 and the signal processing unit 22 function as an image generation unit under the control of the control unit 28.

  In this monitoring operation, the photographed image is only displayed on the liquid crystal monitor (LCD) 9 functioning as an electronic viewfinder, and the release button 2 is not yet pressed (including half-pressed).

By displaying the photographed image on the liquid crystal monitor (LCD) 9, the photographer can check the composition for photographing the still image. Note that the digital camera 1 can also output a TV video signal from the display output control unit 38 and display a captured image on an external TV (TV) (not shown) via a video cable.
(Defective pixel and its detection method)
In an imaging device (such as the CCD 20 described above) used in a digital still camera, not all light receiving elements are formed with the same performance in the semiconductor manufacturing process, and some light receiving elements are defective. May end up. A pixel corresponding to the light receiving element in which this defect has occurred is hereinafter referred to as a defective pixel. However, even if there are a predetermined number or less of defective pixels due to the product yield, it is generally used as a non-defective image sensor in a product such as a digital still camera. For this reason, in the digital camera 1, the ordinate and abscissa (address) of the defective pixel on the CCD 20 (hereinafter referred to as defective pixel data) are stored in advance in the RAM 44 or the like and stored in the RAM 44 or the like in the imaging data at the time of shooting. A defective pixel correction process is performed on the pixel data corresponding to the address of the defective pixel. For this defective pixel, for example, a pure white image having an AE evaluation value that is an average value of the luminance values of the entire image of about 128 is acquired, and in each pixel data of the image data, a difference from the luminance value of 128 is predetermined. It can be detected by judging whether it is larger than the value. Here, those having a luminance value lower than 128 are referred to as black scratches, and those having a luminance value higher than 128 are referred to as white scratches.

In addition to the above-described defective pixels, there are also defective pixels that have poor dark current characteristics (temperature scratches). This temperature flaw is a dynamic defect factor in which the influence increases as the dark current increases or accumulates, that is, as the temperature of the image sensor increases or the exposure time increases. For this reason, a defective pixel having this temperature flaw usually does not cause a problem with exposure of about 1 second, but as the exposure time becomes longer, the increase in dark current increases, which is superimposed on the image data and becomes a bright spot on the image. Looks like. This is what appears as dotted star-like noise in an image obtained by long exposure photography such as night scene photography. There are about 60 defective pixels with temperature flaws with respect to an image sensor with 3 million pixels, and addresses of individual defective pixels can be stored in the RAM 44 or the like. However, since defective pixels with temperature flaws are dynamic defect factors as described above, they must be corrected by imaging parameters such as exposure time. The defective pixel having the temperature flaw is detected by, for example, acquiring a true-dark image by shielding light and determining whether the luminance value is larger than a predetermined value in each pixel data of the image data. Can do.
(Basic correction method for defective pixels)
Correction of defective pixels is basically performed by the following method.

  First, in the digital camera 1, data of defective pixel addresses (vertical and horizontal coordinates on the CCD 20) detected as described above and stored in the RAM 44 or the like is registered in the storage unit 43 of the defective pixel correction unit 42 of the signal processing unit 22. (Store. Thereafter, when the release 2 (see FIG. 1) is pushed down and the image data acquired by the CCD 20 is propagated through the signal processing unit 22 as described above, the image data (image that is the image data of the generation process) is transmitted. Generated data) is sent to the defective pixel correction unit 42. The defective pixel correction unit 42 reads out pixel data corresponding to the address of the defective pixel registered in the storage unit 43 in the image data (each pixel data), and corresponds to the same color as seen by the surrounding RGB filters. Interpolate with pixel data.

  In the case of green (G), the peripheral pixels here refer to both adjacent pixels adjacent in the horizontal direction, adjacent pixels adjacent to each other in the vertical direction, or adjacent pixels adjacent to each other in the diagonal direction, such as blue (B) and In the case of red (R), it means both adjacent pixels adjacent in the horizontal direction or adjacent pixels adjacent in the vertical direction.

  In the first embodiment, basically, interpolation is performed by using the average value of pixel data on both sides adjacent to each other in the horizontal direction of the target defective pixel (defective pixel to be corrected) as pixel data of the target defective pixel. To do. Here, when the noticed defective pixel is located at the left end or the right end in the image data, the pixel data adjacent to the horizontal direction is corrected as the pixel data of the noticed defective pixel.

  Further, in the first embodiment, even when two defective pixels of the same color viewed in the RGB filter are arranged in the horizontal direction (hereinafter also referred to as consecutive two defective pixels), correction can be performed. . When correcting the focused defective pixels of the two consecutive pixels, pixel data in which one defective pixel is adjacent to the other defective pixel on the opposite side is set as pixel data of the one defective pixel, and the other defective pixel is Interpolation is performed by using pixel data adjacent to the defective pixel as the pixel data of the other defective pixel. Here, when the defective pixel of the continuous two pixels of interest is located at the left end or the right end in the image data, the pixel data for two pixels of the pixel data adjacent in the horizontal direction and the pixel data adjacent thereto are Correction is performed by using pixel data of the defective pixels of the two consecutive pixels of interest.

  Here, in the storage unit 43 of the defective pixel correction unit 42, the number of defective pixels (the amount of defective pixel data) that can be registered (the amount of defective pixel data) is finite. For example, the digital camera of the first embodiment 1 is 512. This is because the digital camera 1 is required to reduce the overall size of the device as much as possible as in a general imaging device, and a mechanism for various functions is accommodated therein. The amount of defective pixel data (defective pixels) that cannot be made large-scale, and that can be registered (stored) in the storage unit 43, such as the circuit configuration of the defective pixel correction unit 42 and its storage unit 43, etc. The number is limited. For this reason, in a conventional digital still camera or the like, if there are defective pixels exceeding the number of defective pixels that can be registered (stored) in the storage unit of the defective pixel correction unit in the image sensor, all of them can be corrected for defective pixels. could not.

The present invention makes it possible to correct all defective pixels even when there are defective pixels exceeding the number of defective pixels that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42. Thus, the defective pixel correction processing content executed by the defective pixel correction unit 42 is different from the conventional one. This will be described in detail below.
(Defective pixel correction process)
In the following description, the number of defective pixels that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42 (the registration limit number) is 500 for easy understanding. Further, in order to set the determination criterion for the change in the influence of the defective pixel as the exposure time and the setting condition of the determination criterion as the number of seconds set as the exposure time, when detecting the defective pixel in the CCD 20, the dark image is shielded from light. (Dark image) is acquired (photographed), and the luminance value of a pixel (pixel data) in the acquired image (image data) is set as a defective pixel level, and the luminance value that appears as a bright spot in the acquired image (image data) ( (Defective pixel level) is used as a correction judgment value. In the first embodiment, if the defective pixel level (luminance value) is 240 or more, it appears as a bright spot in the image (image data) when the exposure time is 2 seconds, so that the correction judgment value is 240 to 255, and the defective pixel If the level (luminance value) is 220 or more, it appears as a bright spot in the image (image data) when the exposure time is 4 seconds. Therefore, the correction judgment value is 220 to 239, and the defective pixel level (luminance value) is 200. When the exposure time is 8 seconds, it appears as a bright spot in the image (image data). Therefore, the correction judgment value is 200 to 219, and the exposure time is when the defective pixel level (luminance value) is 180 or more. Since it appears as a bright spot in the image (image data) in the case of 16 seconds, the correction judgment value is set to 180 to 199. In the first embodiment, the higher the defective pixel level (luminance value) is, the higher the correction priority is the defective pixel (temperature scratch). Note that the luminance value as the defective pixel level is an output value with the full scale of the output being 255, and indicates that the larger the value, the larger (higher) output.

  In the digital camera 1 according to the first embodiment, the control unit 28 executes the program stored in the ROM 24 to appropriately control the defective pixel correction unit 42 and the like, thereby performing the defective pixel correction process according to the present invention. In this defective pixel correction process, basically, the number of defective pixels that can be corrected by one defective pixel correction operation in the defective pixel correction unit 42 becomes the number of defective pixels registered (stored) in the storage unit 43. When the number of defective pixels (registration limit number) that can be registered in the storage unit 43 is exceeded, the defective pixel correction operation in the defective pixel correction unit 42 is repeated while rewriting the registration contents in the storage unit 43, thereby obtaining an image sensor. This makes it possible to correct all the defective pixels of the (CCD 20). For this defective pixel correction process, the number of defective pixels to be corrected is set according to the exposure time. Determination of the number of correction pixels according to the exposure time is performed as follows.

  The number of defective pixels depends on the characteristics of the image sensor (sensor). For example, in CMOS, it is said that 10 times more defective pixels appear than CCD even under the same conditions. Therefore, in the state where the sensor (CCD 20 in the first embodiment) is actually mounted on the digital still camera (the digital camera 1 in the first embodiment), the number of defective pixels that appear in the exposure 2 seconds and the defects that appear in the exposure 4 seconds. It is sufficient to statistically grasp the number of defective pixels that appear at each set exposure time, such as the number of pixels, and to correct all the defective pixels that appear according to the exposure time. Is determined as the number of corrected pixels corresponding to the exposure time. For this reason, it can be said that the correction judgment value for the exposure time, that is, each numerical value of the defective pixel level (luminance value) is a guideline for setting the number of correction pixels according to the exposure time. In this defective pixel correction process, defective pixels are corrected by the number of corrected pixels corresponding to the exposure time in order from the pixel with the highest defective pixel level.

  In the first embodiment, the correction pixel number when the exposure time is 2 seconds is 200, the correction pixel number when the exposure time is 4 seconds is 500, and the correction pixel number when the exposure time is 8 seconds is 700. The number of corrected pixels when the exposure time is 16 seconds is 1000 (see FIG. 4). That is, statistically, the number of pixels with a defective pixel level (luminance value) of 240 to 255 is 200 or less, the number of pixels with a defective pixel level (luminance value) of 220 to 239 is 500 or less, and the defective pixel This means that the number of pixels with a level (luminance value) of 200 to 219 is 700 or less, and the number of pixels with a defective pixel level (luminance value) of 180 to 199 is 1000 or less.

  In the first embodiment, the control unit 28 uses the detected defective pixel address (vertical and horizontal coordinates on the CCD 20) as defective pixel data for the defective pixel correction processing with the number of corrected pixels corresponding to the set exposure time. Data is stored in the RAM 44 as follows in association with each defective pixel level and the set exposure time. In the first embodiment, defective pixel data to be appropriately registered in the storage unit 43 is stored in the RAM 44 as will be described later. However, the memory in which the control unit 28 can appropriately store and read data is used. There is no limitation to the first embodiment.

First, as shown in FIG. 4, from among all the pixels of the CCD 20, to detect the 200 pixels in descending order of pixels defective pixel level as a defective pixel, the first address of the defective pixel of the priority correction pixel group P ₁ The data is stored in the RAM 44 as data. 200 defective pixel detected as the first priority correction pixel group P ₁ is, as it becomes a 200 defective pixel correcting when the exposure time is 2 seconds.

Next, among all the pixels of the CCD 20, 500 pixels are detected in order from the pixel having the highest defective pixel level as a defective pixel, and 200 of the first priority correction pixel group P ₁ is detected from the detected 500 pixels. elect minus the number of defective pixels as defective pixels, as the address data of the defective pixel of the second priority correction pixel group P _2, and stores it in the RAM 44. This is because, for example, in the process of detecting 500 pixels in order from the pixel having the highest defective pixel level among all the pixels of the CCD 20, among the 200 defective pixels in the first priority correction pixel group P ₁ among them. This is done by selecting a defective pixel level lower than the lowest defective pixel level. For this reason, 200 defective pixels of the first priority correction pixel group P ₁ are selected from 500 pixels having a high defective pixel level among all the pixels of the CCD ₂₀ as defective pixels of the second priority correction pixel group P _2. The remaining 300 pixels after subtraction are detected. 300 pieces of defective pixels detected as the second priority correction pixel group P ₂ by adding the first 200 defective pixel of the priority correction pixel group P _1, the exposure time is corrected in the case of 4 seconds There are 500 defective pixels. Further, in the defective pixel correction processing according to the present invention, the number of defective pixels that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42 is determined in order of correction priority from the highest defective pixel level, as will be described later. Since the number of defective pixels that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42 is 500 in the first embodiment because it is set as a high defective pixel, the correction is performed when the exposure time is 4 seconds. The defective pixels directly become defective pixels having a high correction priority.

Next, 700 pixels are detected as defective pixels in order from the pixel having the highest defective pixel level among all the pixels of the CCD 20, and 200 of the first priority correction pixel group P ₁ is detected from the detected 700 pixels. the minus the 300 pieces of defective pixel number of defective pixels and second priority correction pixel group P ₂ is selected as the defective pixel, as the address data of the defective pixel of the third priority correction pixel group P _3, and stored in RAM44 . This is because, for example, in the process of detecting 700 pixels in order from the pixel with the highest defective pixel level among all the pixels of the CCD 20, among the 300 defective pixels in the second priority correction pixel group P ₂ among them. This is done by selecting a defective pixel level lower than the lowest defective pixel level. For this reason, as defective pixels in the third priority correction pixel group P ₃ , 200 defective pixels in the first priority correction pixel group P ₁ , from 700 pixels having a high defective pixel level among all the pixels of the CCD 20, and the second 300 of the defective pixel of the priority correction pixel group P ₂ is subtracted, and thus to detect the remaining 200 pixels. The third priority 200 defective pixels detected as the correction pixel group P ₃ is 200 300 defective pixel of the defective pixel and a second priority correction pixel group P ₂ of the first priority correction pixel group P ₁ By adding, 700 defective pixels are corrected when the exposure time is 8 seconds.

Next, 1000 pixels are detected as defective pixels in order from the pixel having the highest defective pixel level among all the pixels of the CCD 20, and 200 of the first priority correction pixel group P ₁ is detected from the detected 1000 pixels. elected number of defective pixels, the minus the 200 defective pixel of the second priority 300 of the defective pixel of the corrected pixel group P ₂ and the third priority correction pixel group P ₃ as a defective pixel, the fourth priority corrected pixel as the address data of the defective pixel group _{P 4,} and stored in RAM 44. This, for example, from among all the pixels of the CCD 20, in the process of the defective pixel level to detect the 1000 pixels from a high pixel sequentially, in a third 200 defective pixel of the priority correction pixel group P ₃ of them This is done by selecting a defective pixel level lower than the lowest defective pixel level. For this reason, as defective pixels of the fourth priority correction pixel group P ₄ , from 200 pixels having a high defective pixel level among all the pixels of the CCD 20, 200 defective pixels of the first priority correction pixel group P ₁ , the 200 defective pixel of the second priority 300 of the defective pixel of the corrected pixel group P ₂ and the third priority correction pixel group P ₃ is subtracted, and thus to detect the remaining 300 pixels. The 300 pieces of defective pixels detected as the fourth priority correction pixel group P ₄ is 200 defective pixel of the first priority correction pixel group P _1, 300 pieces of defective pixels of the second priority correction pixel group P ₂ and By adding 200 defective pixels of the third priority correction pixel group P ₃ , 1000 defective pixels are corrected when the exposure time is 16 seconds.

Since the number of correction pixels is set as described above, in each defective pixel of each priority correction pixel group P _N (N = 1 to 4), among all the pixels of the CCD 20, the correction judgment value, Defective pixels that contain all of the defective pixel level pixels within the range of the defective pixel level (luminance value) as a standard for setting without leaking, but are smaller than the defective pixel level (luminance value) as a standard for setting It is possible that level pixels are included.

The control unit 28 according to the first embodiment uses the address data (see FIG. 4) of each priority correction pixel group P _N (N = 1 to 4) stored in the RAM 44 as described above, that is, uses each defective pixel data to detect a defect. Perform pixel correction processing. FIG. 5 is a flowchart showing details of the defective pixel correction processing of the first embodiment executed by the control unit 28, and FIG. 6 is a flowchart showing processing details of step S4 in the flowchart of FIG. Hereinafter, each step of the flowchart of FIG. 5 (including each step of the flowchart of FIG. 6) will be described with reference to FIG.

  In step S1, the set exposure time is acquired, and the process proceeds to step S2. In this step S1, when the release 2 is pushed down and the photographing operation is started, information on the exposure time at the time of photographing that can be obtained by this is acquired. The exposure time may be set by the control unit 28 depending on the shooting mode, the atmosphere at the shooting site, or the like, or may be set by operating the operation unit 41 (see FIG. 2). Since the exposure time at the time of shooting is determined when it is started, information on the exposure time is acquired with the start of the shooting operation.

  In Step S2, following the acquisition of the set exposure time in Step S1, it is determined whether or not the acquired exposure time is 8 seconds or more. If Yes, the process proceeds to Step S3. If No, Step S4 is performed. Proceed to In step S2, the length (numerical value) of the exposure time acquired in step S1 is determined in order to switch the content of the defective pixel correction process, that is, to switch the number of defective pixel correction operations by the defective pixel correction unit 42. Yes. As described above, this is because the number of correction pixels corresponding to the exposure time is preset and the number of defective pixels (registration limit number) that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42. Depends on the limits. In the first embodiment, the number of defective pixels that can be registered in the storage unit 43 (registration limit number) is 500, whereas the number of corrected pixels when the exposure time is 4 seconds is set to 500 and the exposure time is 8 seconds. In this case, since the number of corrected pixels is set to 700, the number of corrected pixels set to have an exposure time of 8 seconds or more exceeds the registration limit number. The process proceeds to step S3 so that the exposure time is 4 seconds or less, and the number of corrected pixels set to be less than the registration limit number is less than the registration limit number, so that the defective pixel correction operation by the defective pixel correction unit 42 is performed once. Proceed to step S4.

In step S3, following the determination that the exposure time in step S2 is 8 seconds or longer, correction of defective pixels having a high correction priority is executed, and the flow proceeds to step S4. In this step S3, since it is determined in step S2 that the exposure time is 8 seconds or more, there are more defective pixels than the number of defective pixels (registration limit number) that can be registered in the storage unit 43 of the defective pixel correction unit 42. Therefore, the defective pixels corresponding to the registration limit number in the storage unit 43 are corrected in the descending order of correction priority, that is, in the descending order of defective pixel level (luminance value). In Example 1, since the registration limit number of storage unit 43 is 500, the address data of the first 200 defective pixel of the priority correction pixel group P _1, the second priority correction pixel group P ₂ 300 pieces of The defective pixel address data is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42. Thereafter, the defective pixel correction unit 42 reads out pixel data corresponding to the address of the defective pixel registered in the storage unit 43 in the transmitted image data (each pixel data thereof), and looks at the surrounding RGB filters. By interpolating with pixel data corresponding to the same color, a total of 500 pixels including 200 defective pixels in the first priority correction pixel group P ₁ and 300 defective pixels in the second priority correction pixel group P ₂ are obtained. The defective pixel is corrected. Here, since all of the address data of the 500 defective pixels can be registered in the storage unit 43, the defective pixel correction unit 42 performs the defective pixel correction operation once, and the 500 correction pixels having a high correction priority. All defective pixels can be corrected. If the number of defective pixels in each priority correction pixel group P _N (N = 1 to 4) does not correspond to the registration limit number in the storage unit 43, registration can be performed in order from the higher correction priority. It is desirable to store in advance the address data of the registration limit number of defective pixels in the RAM 44 so as to enable registration in the storage unit 43.

In step S4, following the determination that the exposure time in step S2 is not more than 8 seconds or the execution of correction of defective pixels having a high correction priority in step S3, preparation for defective pixel correction according to the exposure time is performed. To go to step S5. In step S4, address data of each priority correction pixel group P _N (N = 1 to 4) is appropriately read out from the RAM 44 in order to correct the defective pixels having the correction pixel number corresponding to the exposure time acquired in step S1. By registering (storing) in the storage unit 43 of the defective pixel correction unit 42, preparation for defective pixel correction corresponding to the exposure time is performed (this is referred to as process A). That is, a substantial correction pixel number is set. Below, each step of the flowchart of FIG. 6 which shows the processing content in the preparation (step S4) of defect pixel correction | amendment according to this exposure time is demonstrated.

  In step S11, it is determined whether or not the exposure time acquired in step S1 is 2 seconds. If Yes, the process proceeds to step S12. If No, the process proceeds to step S13.

In step S12, following the determination that the exposure time in step S11 is 2 seconds, preparation for defective pixel correction corresponding to the exposure time of 2 seconds is made, and the process A (step S4) is terminated, and step S5 is completed. Proceed to (see FIG. 5). Here, when the exposure time is 2 seconds, in the flowchart of FIG. 5, the process proceeds from step S1 to step S2 to step S4. In the step S12, preparation of the defective pixel correction corresponding to the 2-second exposure time, the address data 200 of the defective pixels of the first priority correction pixel group P ₁ (see FIG. 4) is read from the RAM 44, the defective pixel Register (store) in the storage unit 43 of the correction unit 42.

  In step S13, following the determination that the exposure time in step S11 is not 2 seconds, it is determined whether or not the exposure time acquired in step S1 is 4 seconds. If Yes, the process proceeds to step S14. In this case, the process proceeds to step S15.

In step S14, following the determination that the exposure time in step S13 is 4 seconds, preparation for defective pixel correction according to the exposure time of 4 seconds is made, and the process A (step S4) is terminated, and step S5 is completed. Proceed to (see FIG. 5). Here, when the exposure time is 4 seconds, in the flowchart of FIG. 5, it is a scene in which the process proceeds from step S1 to step S2 to step S4. In the step S14, preparation of the defective pixel correction corresponding to 4 seconds exposure time, the address data and 300 of the second priority correction pixel group P ₂ of the first 200 defective pixel of the priority correction pixel group P ₁ The defective pixel address data (see FIG. 4) is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

  In step S15, following the determination that the exposure time in step S13 is not 4 seconds, it is determined whether or not the exposure time acquired in step S1 is 8 seconds. If Yes, the process proceeds to step S16. In this case, the process proceeds to step S17.

In step S16, following the determination in step S15 that the exposure time is 8 seconds, preparation for defective pixel correction corresponding to the exposure time of 8 seconds is made, and processing A (step S4) is terminated, and step S5 is completed. Proceed to (see FIG. 5). Here, when the exposure time is 8 seconds, 500 defects having a high correction priority are obtained in the flowchart of FIG. 5 because the process proceeds from step S1 to step S2 to step S3 to step S4. pixels (the 200 defective pixel of the first priority correction pixel group P _1, and 300 pieces of defective pixels of the second priority correction pixel group P _2, a total of 500 defective pixel) correction has already been completed . Therefore, in step S16, preparation of the defective pixel correction corresponding to 8 seconds exposure time, 200 defective pixel address data of the third priority correction pixel group P ₃ (see FIG. 4) is read from the RAM 44, Register (store) in the storage unit 43 of the defective pixel correction unit 42.

In step S17, following the determination that the exposure time in step S15 is not 8 seconds, preparation for defective pixel correction corresponding to the exposure time of 16 seconds is performed, and the process A (step S4) is terminated, and step S5 is completed. Proceed to (see FIG. 5). Here, if the exposure time is not 8 seconds in step S15, it is set to 16 seconds which is the upper limit of the set exposure time. In this case, in the flowchart of FIG. 5, since the scene proceeds from step S 1 → step S 2 → step S 3 → step S 4, 500 defective pixels having a high correction priority (first priority correction pixel group P ₁ in total 200 defective pixels and 300 defective pixels in the second priority correction pixel group P ₂ , a total of 500 defective pixels) has already been corrected. Therefore, in step S17, preparation of the defective pixel correction corresponding to 16 seconds of exposure time, 300 of the third preferred address data of 200 defective pixel correction pixel group P ₃ and fourth priority correction pixel group P ₄ The address data (see FIG. 4) of the defective pixels is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

In step S5, following the preparation of defective pixel correction corresponding to the exposure time in step S4, defective pixel correction corresponding to the exposure time is executed, and the flowchart of FIG. 5 ends. In step S5, the defective pixel correction unit 42 reads out pixel data corresponding to the address of the defective pixel registered in the storage unit 43 in the transmitted image data (each pixel data thereof), and the surrounding RGB filters. In step S4, the number of corrected pixels is corrected according to the preparation for defective pixel correction according to the exposure time in step S4. Specifically, when the exposure time is 2 seconds, 200 defective pixels of the first priority correction pixel group P ₁ are corrected, and when the exposure time is 4 seconds, 200 pixels of the first priority correction pixel group P ₁ are corrected. performs the defective pixel and a total of 500 correcting defective pixels of the second 300 of the defective pixel of the priority correction pixel group P _2, when the exposure time is 8 seconds, 200 of the third priority correction pixel group P ₃ perform the correction of the defective pixel, when the exposure time is 16 seconds, the third 200 and the total of 500 defective pixel with 300 defective pixel of the fourth priority correction pixel group P ₄ priority correction pixel group P ₃ Perform the correction. At this time, since any defective pixel address data can be registered in the storage unit 43 in any scene, the defective pixel correction unit 42 performs correction according to the exposure time in one defective pixel correction operation. All of the defective pixels of the number of pixels can be corrected. For this reason, in step S5, when the exposure time is 2 seconds or 4 seconds, the correction of the number of correction pixels according to the exposure time acquired in step S1 is performed as the defective pixel correction operation in the first defective pixel correction unit 42. When the exposure time is 8 seconds or 16 seconds, the defective pixel correction operation in the second defective pixel correction unit 42 following step S3 is performed from the correction pixel number corresponding to the exposure time acquired in step S1. Then, the correction of the remaining correction pixel number obtained by subtracting the correction pixel number 500 of the defective pixel correction executed in step S3 is executed without overlapping the pixel data to be corrected.
(Defective pixel correction processing in high-speed shooting processing)
Here, when a high-speed shooting process is necessary (for example, continuous shooting or the like) depending on the shooting mode, the atmosphere at the shooting site, or the like, the defective pixel correction unit 42 repeatedly performs the above-described defective pixel correction process (described above embodiment). If the defective pixel correction operation is performed twice, the processing time for the defective pixel correction becomes a problem. For this reason, in the defective pixel correction processing according to the present invention, when high-speed shooting processing such as continuous shooting is required, it is assumed that pixels with a high defective pixel level (defective pixels with a high correction priority) are corrected with priority. As described above, the defective pixel correction operation in the defective pixel correction unit 42 is performed only once. This is because a pixel having a high defective pixel level is a pixel that tends to appear in an image (image data) as a defective pixel. In this case, for example, when the defective pixel correction operation is executed in step S3 in the flowchart of FIG. 5, the defective pixel correction process is terminated without proceeding to step S4. Accordingly, when high-speed shooting processing such as continuous shooting is required, it is possible to correct defective pixels that are likely to appear in an image (image data) while reducing the processing time required for defective pixel correction.
(Operation of defective pixel correction processing of the present invention)
In the defective pixel correction process of the present invention, as described above, when the number of defective pixels to be corrected in the imaging device (CCD 20) exceeds the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42. Since the operation of correcting defective pixels in the defective pixel correction unit 42 is repeated while rewriting the registration content in the storage unit 43, that is, changing the registration content of defective pixel data in the storage unit 43, the imaging element (CCD 20 ) All defective pixels can be corrected. Specifically, in the first embodiment, the registration limit number that can be registered (stored) in the storage unit 43 is 500, but the number of correction pixels necessary according to the exposure time is set to be 2 seconds. The correction pixel number is 200, the correction pixel number when the exposure time is 4 seconds is 500, the correction pixel number when the exposure time is 8 seconds is 700, and the correction pixel number when the exposure time is 16 seconds is 1000. Therefore, when the exposure time is 2 seconds or 4 seconds, the correction of the number of correction pixels according to the exposure time acquired in step S1 is performed as the defective pixel correction operation in the defective pixel correction unit 42. When the exposure time is 8 seconds or 16 seconds, the defect pixel correction operation in the first defective pixel correction unit 42 is performed as a defective pixel correction operation for the first time in step S3, and 500 defective pixels having a high correction priority are corrected. Second defect following S3 As a defective pixel correction operation in the elementary correction unit 42, correction of the remaining correction pixel number by subtracting the correction pixel number 500 of the defective pixel correction executed in step S3 from the correction pixel number corresponding to the exposure time acquired in step S1. Is executed in step S5. Thus, even when the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42 is exceeded (in the first embodiment, the exposure time is 8 seconds or 16 seconds), Correction is performed by the defective pixel correction operation in the defective pixel correction unit 42 for the second time (multiple times depending on setting conditions), so regardless of the relationship between the number of defective pixels that need to be corrected and the registration limit number of the storage unit 43. All defective pixels of the image sensor (CCD 20) can be corrected.

  Further, since the number of pixels to be corrected is changed according to the exposure time, it is possible to correct only the necessary number of correction pixels according to the exposure time, in other words, it is possible to appropriately correct without excess or deficiency, It is possible to prevent image quality deterioration due to insufficient correction or overcorrection of defective pixels. This is because if defective pixels appearing in the image (image data) are not corrected due to insufficient correction, the defective pixels may be recognized on the image. If a non-appearing pixel is corrected as a defective pixel, the spatial resolution is obtained by substantially smoothing the periphery of the pixel even though it is not a problem pixel (and its periphery) in the image before correction. This is because there is a risk of causing deterioration or the like.

  Furthermore, since pixels with a high defective pixel level are preferentially corrected, pixels that are easily recognized as defective pixels on the image (image data) can be preferentially corrected, so that the image quality can be improved more efficiently. be able to.

  For this reason, in the digital camera 1 of the imaging device according to the present invention, it is possible to appropriately correct all defective pixels to be corrected regardless of the registration limit number in the storage unit 43. As a result, even when dark current increases or accumulates, for example, an image (image data) taken in a situation where the temperature of the image sensor (CCD 20) rises or the exposure time is long, the quality is improved. Can be made.

  In the first embodiment described above, an example is shown in which the exposure criterion is used as a criterion for determining the change in the influence of defective pixels, and the number of correction pixels corresponding to the exposure time is set. Any method may be used as long as the number of correction pixels is set based on a factor that changes the influence of scratches as a criterion, and is not limited to the first embodiment. For example, the defective pixel correction process can be performed based on the setting of the number of correction pixels according to the ISO sensitivity. In this case, as with the exposure time, the number of defective pixels appearing with ISO sensitivity 100 in the state where the sensor (CCD 20 in Example 1) is actually mounted on the digital still camera (Digital camera 1 in Example 1), ISO Statistically grasp the number of defective pixels that appear at each set ISO sensitivity, such as the number of defective pixels that appear at a sensitivity of 200, and so on, and identify all defective pixels that appear according to the ISO sensitivity. A number that can be determined to be sufficient for correction may be set as the number of corrected pixels according to ISO sensitivity. As an example of this, the number of correction pixels with an exposure time of ISO 100 is 200, the number of correction pixels with ISO 200 is 500, the number of correction pixels with ISO 400 is 700, and the number of correction pixels with ISO 800 is 1000. Thus, defective pixel correction processing can be performed in the same manner as in the case where the number of correction pixels is set corresponding to the exposure time described above. Even in this case, the control unit 28 uses the data of the detected defective pixel address (vertical and horizontal coordinates on the CCD 20) for the defective pixel correction processing with the number of corrected pixels corresponding to the set ISO sensitivity. Stored in the RAM 44.

  In the first embodiment, the number of corrected pixels when the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correcting unit 42 is 500 and the setting of the number of corrected pixels according to the exposure time is 16 seconds is used. Although 1000 was set as the maximum, each numerical value is an example and is not limited to these. In other words, when there are more exposure times, the number of defective pixels corresponding to them may be set as appropriate, and the number of defective pixels that can be registered in the storage unit 43 and the number of correction pixels for the exposure time are also set in the imaging apparatus. It changes according to each setting. Here, when the maximum value of the correction pixel number setting corresponding to the exposure time exceeds twice the number of defective pixels (registration limit number) that can be registered in the storage unit 43, the first defect in step S3 The registration limit number is corrected as a defective pixel correction operation in the pixel correction unit 42, and the defect pixel correction operation in the second defective pixel correction unit 42 following step S3 is performed according to the exposure time acquired in step S1. Defects with a high correction priority corresponding to the registration limit number among the pixels that are excluded from correction of each pixel corresponding to the correction pixel number of the defective pixel correction executed in step S3 from each pixel corresponding to the correction pixel number As a defective pixel correction operation in the defective pixel correction unit 42 for the third time after executing the pixel correction, the defective pixel correction performed in the above-described two times from the number of corrected pixels corresponding to the exposure time acquired in step S1. Complement of The defective pixel correction operation of the registration limit number of pixels in the defective pixel correction unit 42 is performed (for example, by performing correction of the remaining correction pixel number by subtracting the pixel number (registration limit number × 2)). After repeating the correction pixel number) / registration limit number), the remaining correction pixel number may be corrected.

Next, an image pickup apparatus according to Embodiment 2 of the present invention and defective pixel correction processing performed therein will be described. The second embodiment is an example in which part of the processing content in the defective pixel correction process is different from the first embodiment. Since the basic configuration of the image pickup apparatus according to the second embodiment is the same as that of the digital camera 1 that is the image pickup apparatus according to the first embodiment, the same reference numerals are given to portions having the same configuration, and detailed description thereof will be given. Is omitted. FIG. 7 is a description similar to FIG. 4 showing the number of defective pixel addresses (the number of horizontal and vertical coordinates on the CCD 20) stored in the RAM 44 in each priority correction pixel group P _NM in a table. FIG. FIGS. 8A and 8B are explanatory diagrams for explaining a state where the acquired image (image data) is divided into S equal parts. FIG. 8A shows a state of being divided into four parts, and FIG. 8B shows a state of being divided into two parts. .

  The defective pixel correction process according to the second embodiment corrects all the defective pixels even when the number of defective pixels that need to be corrected exceeds the number of defective pixels that can be registered in the storage unit 43 (registration limit number). In this case, the processing time is prevented from increasing as the defective pixel correction unit 42 repeats the defective pixel correction operation. For example, in the digital camera 1 according to the first embodiment, when 10 MB of image data is input to the defective pixel correction unit 42 as a block that performs a defective pixel correction process using hardware, the defective pixel correction of the image data takes about 80 ms. It is assumed that processing time is required, and 1500 defective pixels are corrected with respect to 500 defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42. Then, in the defective pixel correction unit 42, first, correction of 500 defective pixels is performed on the 10 MB image data in order from the pixel having the highest defective pixel level, and then the 10 MB image data is already corrected. 500 defective pixels are corrected in order from the pixel with the highest defective pixel level except for the corrected 500 pixels, and finally the remaining 500 defective pixels are corrected for 10 MB of image data. To do. From this, in order to correct all defective pixels that need to be corrected, the defective pixel correction unit 42 performs processing for three defective pixel correction operations on the 10 MB image data (about 80 ms × 3 = about 240 ms).

  In the defective pixel correction process according to the second embodiment, in the imaging device (CCD 20), when there are defective pixels exceeding the number of defective pixels that can be registered (stored) in the storage unit of the defective pixel correction unit (the set number of corrected pixels is registered). Even when the number exceeds the limit number), it is possible to correct all of the defective pixels, while suppressing an increase in processing time required for the defective pixel correction.

  First, for the defective pixel correction process of the second embodiment, as in the first embodiment, the number of defective pixels appearing at each set exposure time is statistically grasped, and the defective pixels appearing according to the exposure time. The number that can be determined to be sufficient to correct all of the above is set as the number of corrected pixels according to the exposure time. In the second embodiment, the correction pixel number when the exposure time is 2 seconds is 200, the correction pixel number when the exposure time is 4 seconds is 500, and the correction pixel number when the exposure time is 8 seconds is 800. The number of corrected pixels when the exposure time is 16 seconds is 1200, and the number of corrected pixels when the exposure time is 32 seconds is 2000 (see FIG. 7). The number of corrected pixels corresponding to the exposure time is stored in the RAM 44.

  Here, since the maximum value of the correction pixel number corresponding to the exposure time is set to 2000, the control unit 28 sets the number of defective pixels that can be registered in the storage unit 43 for all the 2000 defective pixels. In order to correct by the defective pixel correction unit 42 which is 500, it is necessary to perform the defective pixel correction operation four times (2000/500), and the maximum value of the division number S is set to 4. As shown in FIG. 8, the division number S sets a target area, that is, a data amount in one defective pixel correction operation in the defective pixel correction unit 42 in one image data acquired by the CCD 20. Therefore, the number of the image data is equally divided and is set to any one of 1, 2, and 4 in the second embodiment. This division equally divides the image data (data amount) so that defective pixel correction by the defective pixel correction unit 42 is possible. In the second embodiment, the defective pixel correction operation in the defective pixel correction unit 42 basically uses the average value of the adjacent pixel data in the horizontal direction of the defective pixel of interest as the pixel data of the defective pixel of interest. Therefore, as shown in FIG. 8, the image data is equally divided in the vertical direction so that the equal lines in the image data are in the horizontal direction.

  In the second embodiment, the control unit 28 converts the data of the detected defective pixel address (vertical and horizontal coordinates on the CCD 20) into each of the defective pixel correction processing with the number of corrected pixels corresponding to the set exposure time. In association with the defective pixel level and the set exposure time, it is stored in the RAM 44 as follows.

The basic method for detecting defective pixels in the second embodiment is the same as that in the first embodiment. However, as described above, since the maximum value of the number of divisions S is set to 4, an image divided into four areas is used. For each data (hereinafter, divided image data I ₄₁ (divided image A), divided image data I ₄₂ (divided image B), divided image data I ₄₃ (divided image C), divided image data I ₄₄ (divided image D), which will be described later. (See FIG. 8A)), each priority correction pixel group _PNM is acquired. Here, N represents a priority number in order from the highest defective pixel level, that is, represents a priority of each group to be corrected in each image data. The Nth priority correction pixel group in the first embodiment. This corresponds to _N in PN. In the second embodiment, there are five exposure time settings of 2, 4, 8, 16, and 32 seconds, so that 1, 2, 3, 4, and 5 are applied in order. M represents which of the divided images A, B, C, and D divided into four in each N-th group, and 1 in the order of A, B, C, and D. 2, 3, 4 are fitted.

First, as shown in FIG. 7, in each of the divided images A, B, C, and D (see FIG. 8A) in which all the pixels of the CCD 20 are divided into four regions, from among these pixels, 50 pixels in the order of the defective pixel level from a high pixel detected as a defective pixel, the address data of the defective pixel of the first priority correction pixel group P ₁₁ in the divided image a, the first priority correction pixel in the divided image B and the address data of the defective pixel group P _12, and the address data of the defective pixel of the first priority correction pixel group P ₁₃ in the divided image C, the address of the defective pixel of the first priority correction pixel group P ₁₄ in the divided image D The data is stored in the RAM 44 as data. Here, the 50 pieces of defective pixels detected as a first priority correction pixel group P _11, and 50 defective pixels detected as a first priority correction pixel group P _12, as a first priority correction pixel group P ₁₃ A combination of the 50 defective pixels detected and the 50 defective pixels detected as the first priority correction pixel group P ₁₄ includes 200 defective pixels to be corrected when the exposure time is 2 seconds. It becomes (first priority correction pixels is defined as _{P 1).}

Next, 125 pixels are detected as defective pixels in order from the pixel having the highest defective pixel level from among the pixels of each of the divided images A, B, C, and D, and corresponding from the detected 125 pixels. The first priority correction pixel group P _1M (M = 1 to 4) except 50 defective pixels is selected as a defective pixel, and the second priority correction pixel group P _2M (M = 1 to 4) is defective. It is stored in the RAM 44 as pixel address data. These 75 defective pixels detected as the second priority correction pixel group P ₂₁ , 75 defective pixels detected as the second priority correction pixel group P ₂₂ , and the second priority correction pixel group P ₂₃ A combination of the 75 defective pixels detected and the 75 defective pixels detected as the second priority correction pixel group P ₂₄ (referred to as the second priority correction pixel group P ₂ ) is the first priority described above. by adding 200 defective pixel of the priority correction pixel group P _1, a 500 defective pixel exposure time is corrected in the case of 4 seconds. Further, in the defective pixel correction processing according to the present invention, the number of defective pixels that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42 is set as a defective pixel having a higher correction priority in order from the highest defective pixel level. Therefore, in Example 2, since the number of defective pixels that can be registered (stored) in the storage unit 43 of the defective pixel correction unit 42 is 500, 500 defective pixels to be corrected when the exposure time is 4 seconds. As such, it becomes a defective pixel having a high correction priority. As in the first embodiment, this defective pixel having a high correction priority is used when the defective pixel correction unit 42 performs the defective pixel correction operation only once when high-speed shooting processing such as continuous shooting is required.

Next, 200 pixels are detected as defective pixels in order from the pixel having the highest defective pixel level among the pixels of each of the divided images A, B, C, and D, and the corresponding pixels are detected from the detected 200 pixels. Excluding 50 defective pixels in each first priority correction pixel group P _1M (M = 1 to 4) and 75 defective pixels in each second priority correction pixel group P _2M (M = 1 to 4) The defective pixel is selected and stored in the RAM 44 as address data of the defective pixel of the third priority correction pixel group P _3M (M = 1 to 4). These 75 defective pixels detected as the third priority correction pixel group P ₃₁ , 75 defective pixels detected as the third priority correction pixel group P ₃₂ , and the third priority correction pixel group P ₃₃ A combination of the 75 defective pixels detected and the 75 defective pixels detected as the third priority correction pixel group P ₃₄ (referred to as a third priority correction pixel group P ₃ ) is the first one described above. by adding 200 300 of the defective pixels of the defective pixel and a second priority correction pixel group P ₂ priority correction pixel group P _1, a 800 amino defective pixel exposure time is corrected in the case of 8 seconds .

Next, among the pixels of each of the divided images A, B, C, and D, 300 pixels are detected as defective pixels in order from the pixel having the highest defective pixel level, and the corresponding pixels are detected from the detected 300 pixels. 50 defective pixels in each first priority correction pixel group P _1M (M = 1 to 4), 75 defective pixels in each second priority correction pixel group P _2M (M = 1 to 4), and each third priority A pixel excluding 75 defective pixels in the correction pixel group P _3M (M = 1 to 4) is selected as a defective pixel, and address data of defective pixels in the fourth priority correction pixel group P _4M (M = 1 to 4). Is stored in the RAM 44. These 100 defective pixels detected as the fourth priority correction pixel group P ₄₁ , 100 defective pixels detected as the fourth priority correction pixel group P ₄₂ , and the fourth priority correction pixel group P ₄₃ A combination of the 100 defective pixels detected and the 100 defective pixels detected as the fourth priority correction pixel group P ₄₄ (referred to as the fourth priority correction pixel group P ₄ ) is the first one described above. 200 defective pixel of the priority correction pixel group P _1, by adding 300 of the defective pixel of the second priority correction pixel group P ₂ of 300 defective pixel, and the third priority correction pixel group P _3, the exposure time When there are 16 seconds, 1200 defective pixels are corrected.

Next, among the pixels of each of the divided images A, B, C, and D, 500 pixels are detected as defective pixels in order from the pixel having the highest defective pixel level, and the corresponding pixels are detected from the detected 500 pixels. 50 defective pixels of each first priority correction pixel group P _1M (M = 1 to 4), 75 defective pixels of each second priority correction pixel group P _2M (M = 1 to 4), each third priority A pixel excluding 75 defective pixels of the correction pixel group P _3M (M = 1 to 4) and 100 defective pixels of each fourth priority correction pixel group P _4M (M = 1 to 4) is selected as a defective pixel. Then, it is stored in the RAM 44 as address data of defective pixels of the fifth priority correction pixel group P _5M (M = 1 to 4). These, and 200 defective pixels detected as the fifth priority correction pixel group P _51, and 200 of the defective pixels detected as the fifth priority correction pixel group P _52, as the fifth priority correction pixel group P ₅₃ A combination of the detected 200 defective pixels and the 200 defective pixels detected as the fifth priority correction pixel group P ₅₄ (referred to as a fifth priority correction pixel group P ₅ ) is the first one described above. 200 defective pixel of the priority correction pixel group P _1, 300 pieces of defective pixels of the second priority correction pixel group P _2, the third priority 300 of the defective pixel of the corrected pixel group P ₃ and the fourth priority correction pixel group P By adding 400 defective pixels of ₄ , 2000 defective pixels are corrected when the exposure time is 32 seconds.

In the control unit 28 according to the second embodiment, as described above, address data (see FIG. 7) of each priority correction pixel group P _NM (N = 1 to 5, M = 1 to 4) stored in the RAM 44, that is, each defective pixel data. Is used to perform defective pixel correction processing. FIG. 9 is a flowchart showing details of defective pixel correction processing of the second embodiment executed by the control unit 28, and FIG. 10 is a flowchart showing processing details of step S24 in the flowchart of FIG. Hereinafter, each step of the flowchart of FIG. 9 (including each step of the flowchart of FIG. 10) will be described with reference to FIG.

  In step S21, the set exposure time is acquired, and the process proceeds to step S22. In step S21, as in step S1 of the flowchart of FIG. 5, when the release 2 is pushed down and the photographing operation is started, information on the exposure time at the time of photographing that can be obtained is acquired.

  In step S22, following the acquisition of the exposure time in step S21, the number of corrected pixels corresponding to the acquired exposure time is read, and the process proceeds to step S23. In this step S22, the number of corrected pixels corresponding to the exposure time stored in the RAM 44 is read out.

In step S23, following the readout of the number of corrected pixels corresponding to the exposure time in step S22, a division number S corresponding to the acquired exposure time is set, and the process proceeds to step S24. In this step S23, the number of corrected pixels corresponding to the read exposure time is divided by the number of defective pixels (500) that can be registered in the storage unit 43. If the quotient is 1 or less, the division number S is set to 1. If S is greater than 1 and less than or equal to 2, the division number S is set to 2, and if the quotient is greater than 2, the division number S is set to 4. In other words, the number of defective pixel correction operations required to correct all the correction pixel numbers according to the exposure time by the defective pixel correction unit 42 is set as the division number S (S = 1, 2, 4). Basically, the number of corrected pixels corresponding to the read exposure time is divided by the registration limit number (500 in the second embodiment) in the storage unit 43, and the value obtained by rounding up the decimal point of the quotient is the division number S. However, since the division number S is set to any one of 1, 2, and 4 in the second embodiment, the division number S is set to 4 even when the rounded-up numerical value is 3. To do. Specifically, the division number S when the exposure time is 2 seconds is 1, the division number S when the exposure time is 4 seconds, and the division number S when the exposure time is 8 seconds is 2. The division number S when the exposure time is 16 seconds is 4 and the division number S when the exposure time is 32 seconds is 4 (see FIG. 8A). In step S23, the counter number C is set to 1. The counter number C is equal to each divided image divided by the division number S (A, B, C, and D (see FIG. 8A) when the division number is 4), and E and F when the division number is 2 (FIG. 8). (B))) of the image data (divided image data I _SC (S: number of divisions, C: number of counters)), A, B, C, D ( Alternatively, 1, 2, 3, 4 (or 1, 2) are applied in the order of E, F). Therefore, when the division number S is 2, the image data is divided into the divided image data I ₂₁ of the divided image E and the divided image data I ₂₂ of the divided image F (see FIG. 8B). Is divided image data I ₄₁ of the divided image A, divided image data I ₄₂ of the divided image B, divided image data I ₄₃ of the divided image C, and divided image data I of the divided image D. ₄₄ (see FIG. 8A).

In step S24, following the setting of the division number S in step S23 or the increment of the counter number C in step S27, preparation for defective pixel correction according to the division number S and the exposure time is performed, and the flow proceeds to step S25. move on. In step S24, each of the priority correction pixel groups P _NM (N = 1 to 5) is used to correct the pixels having the correction pixel number corresponding to the exposure time acquired in step S21 (substantially set the correction pixel number). , M = 1 to 4) is appropriately read out from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42 to prepare for defective pixel correction according to the exposure time (this) Process B). In the following, each step of the flowchart of FIG. 10 showing the processing content in preparation for defective pixel correction according to the exposure time (step S24) will be described.

  In step S31, it is determined whether or not the exposure time acquired in step S21 is 2 seconds. If Yes, the process proceeds to step S32. If No, the process proceeds to step S33.

In step S32, following the determination that the exposure time in step S31 is 2 seconds, preparation for defective pixel correction corresponding to the exposure time of 2 seconds is made, and the process B (step S24) is terminated, and step S25 is completed. Proceed to (see FIG. 9). Here, when the exposure time is 2 seconds, the number of divisions S is set to 1 in step S23 of the flowchart of FIG. In the step S32, preparation of the defective pixel correction corresponding to the 2-second exposure time, to correct all of the defective pixel in the divided image data I ₁₁ made directly to the image data since it is divided number 1, first The address data of a total of 200 defective pixels of the address data (see FIG. 7) of 50 defective pixels of the priority correction pixel group P _1M (M = 1 to 4) is read from the RAM 44 and Register (store) in the storage unit 43.

  In step S33, following the determination that the exposure time in step S31 is not 2 seconds, it is determined whether or not the exposure time acquired in step S21 is 4 seconds. If yes, the process proceeds to step S34. In this case, the process proceeds to step S35.

In step S34, following the determination that the exposure time in step S33 is 4 seconds, preparation for defective pixel correction corresponding to the exposure time of 4 seconds is made, and the process B (step S24) is terminated, and step S25 is completed. Proceed to (see FIG. 9). Here, when the exposure time is 4 seconds, the division number S is set to 1 in step S23 of the flowchart of FIG. In the step S34, preparation of the defective pixel correction corresponding to 4 seconds exposure time, to correct all of the defective pixel in the divided image data I ₁₁ made directly to the image data since it is divided number 1, first Address data of 50 defective pixels of the priority correction pixel group P _1M (M = 1 to 4) and address data of 75 defective pixels of the second priority correction pixel group P _2M (M = 1 to 4) ( 7), the address data of 500 defective pixels is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

  In step S35, following the determination that the exposure time in step S33 is not 4 seconds, it is determined whether or not the exposure time acquired in step S21 is 8 seconds. If Yes, the process proceeds to step S36. In this case, the process proceeds to step S39.

  In step S36, following the determination that the exposure time in step S35 is 8 seconds, it is determined whether or not the counter number C is 1, the process proceeds to step S37 if Yes, and in step S38 if No. Proceed to This is a scene where the number of divisions S is set to 2 in step S23 of the flowchart of FIG. 9 when the exposure time is 8 seconds. As will be described later, from step S24 of the flowchart of FIG. In the flow returning to step S24 via step S27, the counter number C is increased from 1 to 2 (step S27), and the defective pixel correction unit 42 for each of the divided image E and the divided image F (see FIG. 8B). This is because the defective pixel correction operation is executed in accordance with the number of counters C and the preparation for defective pixel correction corresponding to the exposure time of 8 seconds differs.

In step S37, following the determination that the counter number C is 1 in step S36, preparation for defective pixel correction of the divided image E (see FIG. 8B) corresponding to an exposure time of 8 seconds is performed. The process B (step S24) is ended, and the process proceeds to step S25 (see FIG. 9). In the step S37, preparation for defective pixel correction of the divided image E corresponding to 8 seconds exposure time (see FIG. 8 (b)), the correction of all defective pixels in the divided image data I ₂₁ to be half of the image data Therefore, a priority number (N = 1 to 3) corresponding to an exposure time of 8 seconds and a number corresponding to a divided image E (divided image A and divided image B) obtained by dividing the image data into two (M = 1, 2), the address data (see FIG. 7) of each defective pixel of the Nth priority correction pixel group P _NM (N = 1 to 3, M = 1 to 2) is read from the RAM 44, and the defective pixel correction unit 42 Is registered (stored) in the storage unit 43. That is, the address data of both 50 defective pixels of the first priority correction pixel group P _1M (1-2), the address data of both 75 defective pixels of the second priority correction pixel group P _2M (1-2), and Address data of a total of 400 defective pixels, that is, address data (see FIG. 7) of both 75 defective pixels of the third priority correction pixel group P _3M (1-2) is read from the RAM 44, and the defective pixel correction unit 42 is read out. Is registered (stored) in the storage unit 43.

In step S38, following the determination that the counter number C in step S36 is not 1, preparation for defective pixel correction of the divided image F (see FIG. 8B) corresponding to an exposure time of 8 seconds is performed. The process B (step S24) is ended, and the process proceeds to step S25 (see FIG. 9). In the step S38, the preparation of the defective pixel correction of the divided image F corresponding to 8 seconds exposure time (see FIG. 8 (b)), the correction of all defective pixels in the divided image data I ₂₂ to be half of the image data Therefore, a priority number (N = 1 to 3) corresponding to an exposure time of 8 seconds and a number corresponding to divided image F (divided image C and divided image D) obtained by dividing the image data into two (M = 3 and 4), the address data (see FIG. 7) of each defective pixel of the N-th priority correction pixel group P _NM (N = 1 to 3, M = 3 to 4) is read from the RAM 44, and the defective pixel correction unit 42 is read out. Is registered (stored) in the storage unit 43. That is, the address data of both 50 defective pixels of the first priority correction pixel group P _1M (3-4), the address data of both 75 defective pixels of the second priority correction pixel group P _2M (3-4), and Address data of a total of 400 defective pixels, that is, address data (see FIG. 7) of both 75 defective pixels of the third priority correction pixel group P _3M (3 to 4) is read from the RAM 44, and the defective pixel correction unit 42 is read out. Is registered (stored) in the storage unit 43.

  In step S39, following the determination that the exposure time in step S35 is not 8 seconds, it is determined whether or not the exposure time acquired in step S21 is 16 seconds. If Yes, the process proceeds to step S40. In this case, the process proceeds to step S41.

In step S40, following the determination that the exposure time in step S39 is 16 seconds, any one of the divided images A, B, C, and D (see FIG. 8A) corresponding to the exposure time of 16 seconds. In step S24, the process proceeds to step S25 (see FIG. 9). This is a scene in which the number of divisions S is set to 4 in step S23 of the flowchart of FIG. 9 when the exposure time is 16 seconds. As will be described later, from step S24 of the flowchart of FIG. In the flow returning to step S24 through step S27, the counter number C is sequentially increased from 1 to 4 (step S27), and defective pixels for each of the divided images A, B, C, and D (see FIG. 8A). Since the correction unit 42 performs the defective pixel correction operation, the preparation for defective pixel correction corresponding to the exposure time of 16 seconds differs depending on the counter number C. In this step S40, 1/4 of the image data is prepared in preparation for correcting defective pixels of any of the divided images A, B, C, and D (see FIG. 8A) corresponding to the exposure time of 16 seconds. In order to correct all defective pixels in the divided image data I _4C (C = 1 to 4), the image data is divided into four with a priority number (N = 1 to 4) corresponding to an exposure time of 16 seconds. N-th priority correction pixel group P _NM (N = 1 to 4, M = 1 to 4) corresponding to the divided images A, B, C, and D (in order M = 1, 2, 3, 4). Are read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42. That is, when the counter number C is 1, for the defective pixel correction of the divided image data I ₄₁ (divided image A), the address data of the 50 defective pixels in the first priority correction pixel group P ₁₁ , the second priority correction 75 pieces of defective pixel address data, third priority correction of 75 pieces of defective pixels of the pixel group P ₃₁ address data and the fourth 100 defective pixel address data of the priority correction pixel group P ₄₁ pixel group P _21, The address data of a total of 300 defective pixels is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42. Further, when the counter number C is 2, because of the defective pixel correction of the divided image data I _{42 (divided} image B), 50 pieces of defective pixel address data of the first priority correction pixel group P _12, the second priority correction 75 pieces of defective pixel address data, third priority correction of 75 pieces of defective pixels of the pixel group P ₃₂ address data and 100 defective pixel address data of the fourth priority correction pixel group P ₄₂ pixel group P _22, The address data of a total of 300 defective pixels is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42. Furthermore, when the counter number C is 3, for the defective pixel correction of the divided image data I ₄₃ (divided image C), the address data of the 50 defective pixels of the first priority correction pixel group P ₁₃ , the second priority correction 75 pieces of defective pixel address data, third priority correction of 75 pieces of defective pixels of the pixel group P ₃₃ address data and 100 defective pixel address data of the fourth priority correction pixel group P ₄₃ pixel group P _23, The address data of a total of 300 defective pixels is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42. Then, when the counter number C is 4, for defective pixel correction of the divided image data I _{44 (divided} image D), 50 pieces of defective pixel address data of the first priority correction pixel group P _14, the second priority correction Address data of 75 defective pixels of the pixel group P ₂₄ , address data of 75 defective pixels of the third priority correction pixel group P ₃₄ , and address data of 100 defective pixels of the fourth priority correction pixel group P ₄₄ , The address data of a total of 300 defective pixels is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42.

In step S41, following the determination that the exposure time in step S39 is not 16 seconds, one of the divided images A, B, C, and D (see FIG. 8A) corresponding to the exposure time of 32 seconds. In step S24, the process proceeds to step S25 (see FIG. 9). This means that if the exposure time is not 16 seconds in step S39, it is set to 32 seconds, which is the upper limit of the set exposure time. In step S23 of the flowchart of FIG. Since it is a set scene, as will be described later, the counter number C is sequentially increased from 1 to 4 in the flow of returning from step S24 to step S27 in the flowchart of FIG. 9 (step S27). ) Since the defective pixel correction unit 42 executes the defective pixel correction operation for each of the divided images A, B, C, and D (see FIG. 8A), the number of counters C corresponds to the exposure time of 32 seconds. This is due to different preparations for defective pixel correction. In this step S41, 1/4 of the image data is prepared as preparation for defective pixel correction of any one of the divided images A, B, C, and D (see FIG. 8A) corresponding to the exposure time of 32 seconds. In order to correct all defective pixels in the divided image data I _4C (C = 1 to 4), the image data is divided into four with a priority number (N = 1 to 5) corresponding to an exposure time of 32 seconds. N-th priority correction pixel group P _NM (N = 1 to 5, M = 1 to 4) corresponding to the divided images A, B, C, and D (in order M = 1, 2, 3, 4). Are read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42. That is, when the counter number C is 1, for the defective pixel correction of the divided image data I ₄₁ (divided image A), the address data of the 50 defective pixels in the first priority correction pixel group P ₁₁ , the second priority correction 75 pieces of defective pixel address data of pixel groups P _21, the third priority 75 of the defective pixel address data for correcting pixel group P _31, 100 pieces of address data of the defective pixel and the fourth priority correction pixel group P ₄₁ A total of 500 defective pixel address data of 200 defective pixels of the fifth priority correction pixel group P ₅₁ are read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42. Further, when the counter number C is 2, because of the defective pixel correction of the divided image data I _{42 (divided} image B), 50 pieces of defective pixel address data of the first priority correction pixel group P _12, the second priority correction Address data of 75 defective pixels of the pixel group P ₂₂ , address data of 75 defective pixels of the third priority correction pixel group P ₃₂ , address data of 100 defective pixels of the fourth priority correction pixel group P ₄₂ , In addition, the address data of 500 defective pixels of the 200 defective pixels of the fifth priority correction pixel group P ₅₂ is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42. Furthermore, when the counter number C is 3, for the defective pixel correction of the divided image data I ₄₃ (divided image C), the address data of the 50 defective pixels of the first priority correction pixel group P ₁₃ , the second priority correction 75 pieces of defective pixel address data of the pixel group P _23, the third priority 75 of the defective pixel address data for correcting pixel group P _33, the address data and of 100 defective pixel of the fourth priority correction pixel group P ₄₃ A total of 500 defective pixel address data of 200 defective pixels in the fifth priority correction pixel group P ₅₃ is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42. Then, when the counter number C is 4, for defective pixel correction of the divided image data I _{44 (divided} image D), 50 pieces of defective pixel address data of the first priority correction pixel group P _14, the second priority correction Address data of 75 defective pixels in the pixel group P ₂₄ , address data of 75 defective pixels in the third priority correction pixel group P ₃₄ , address data of 100 defective pixels in the fourth priority correction pixel group P ₄₄ , and A total of 500 defective pixel address data of 200 defective pixels in the fifth priority correction pixel group P ₅₄ is read from the RAM 44 and registered in the storage unit 43 of the defective pixel correction unit 42.

In step S25, following the preparation of defective pixel correction according to the division number S and the exposure time in step S24, the divided image data I _SC of the divided image corresponding to the counter number C at the division number S according to the exposure time is obtained. On the other hand, defective pixel correction is executed, and the process proceeds to step S26. In the step S25, the defective pixel correction unit 42, the image sent data at (each pixel data) divided image data I _SC corresponding to the counter number C of, registered in the storage unit 43 the defective pixel The pixel data corresponding to the address is read out and interpolated with pixel data corresponding to the same color as seen by the surrounding RGB filters, so that the number of correction pixels corresponding to the defective pixel correction preparation according to the exposure time in step S24 Perform the correction.

In step S26, following the execution of the defective pixel correction for the divided image data I _SC of the divided image corresponding to the counter number C in the divided number S corresponding to the exposure time in step S25, the counter number C is equal to the number of divisions S In the case of Yes, the flowchart of FIG. 9 is ended, and in the case of No, the process proceeds to Step S27. In step S26, by the counter number C is determined whether equal to the division number S corresponding to the exposure time, by Step S25 to perform defective pixel correction on the divided image data I _SC is repeated, It is determined whether or not defective pixel correction has been performed on all divided images, that is, all image data before being divided.

In step S27, following the determination that the counter number C is not equal to the division number S in step S26, the counter number C is incremented (C = C + 1), and the process returns to step S24. In the step S27, in order to perform defective pixel correction for all of the image data before being divided, the number of execution of the defective pixel correction for the divided image data I _SC in step S25 (the counter C) by adding 1 To do.

As described above, in the defective pixel correction process according to the second embodiment, the division number S is set in accordance with the correction pixel number corresponding to the acquired exposure time (step S23), and the image (image data) is divided by the division number S or the like. Thus, defective pixel correction is performed on all of the image data without duplicating the pixel data to be corrected by repeating the defective pixel correction operation in the defective pixel correction unit 42 for each divided image data _ISC . (Step S25). In this case, in step S24, since it has registered address data of each defective pixel in the divided image data I _SC in accordance with the number of executions of the defective pixel correction in the step S25 (counter C) in RAM 44, step S24 By repeating the flow from step S25 to step S26 to step S27 and returning to step S24, defective pixel correction can be performed on all of the image data before being divided. For example, when the exposure time is set to 32 seconds (the number of corrected pixels is 2000), the process proceeds from step S21 to step S22 to step S23, whereby the division number S is set to 4 and the image data is converted. Divide into divided images A, B, C, and D. Thereafter, in step S24, preparation for defective pixel correction of the divided image data I ₄₁ of the divided image A corresponding to the exposure time of 32 seconds is performed, and the process proceeds to step S25, whereby the defective pixel correction unit 42 is sent. In the divided image data I ₄₁ of the obtained image data, pixel data corresponding to the address of the defective pixel registered in the storage unit 43 (50 defective pixels and second priority correction pixels of the first priority correction pixel group P ₁₁ ). 75 defective pixels in group P ₂₁ , 75 defective pixels in third priority correction pixel group P ₃₁ , 100 defective pixels in fourth priority correction pixel group P ₄₁ , and 200 in fifth priority correction pixel group P ₅₁ A total of 500 defective pixels), and interpolating with pixel data corresponding to the same color as seen by the surrounding RGB filters, the exposure time of the divided image A is 32 seconds. The defective pixel correction is completed. Thereafter, the process proceeds to step S26, but since the counter number C is 1 and is not equal to the division number S = 4, the process proceeds to step S27 and returns to step S24, whereby the divided image corresponding to the exposure time of 32 seconds is obtained. Preparation for defective pixel correction of the B divided image data I ₄₂ is performed. Thereafter, by proceeding to step S25, the defective pixel correction unit 42 in the divided image data I ₄₂ of the transmitted image data, pixel data corresponding to the defective pixel address registered in the storage unit 43 ( 50 defective pixels of the first priority correction pixel group _{P 12,} 75 pieces of defective pixels of the second priority correction pixel group _{P 22,} 75 pieces of defective pixels of the third priority correction pixel group _{P 32,} the fourth priority corrected pixel 100 defective pixels in the group P ₄₂ and 200 defective pixels in the fifth priority correction pixel group P ₅₂ , a total of 500 defective pixels), and pixels corresponding to the same color as seen by the surrounding RGB filters By interpolating with the data, the defective pixel correction of the divided image B is completed. When the above processing (the flow from step S24 → step S25 → step S26 → step S27 and returning to step S24) is repeated four times as the division number S, the counter number C is 4 in the fourth step S26. Therefore, the process does not proceed to step S27, and the defective pixel correction of the divided image C and the divided image D is completed in addition to the divided image A and the divided image B described above by ending the flowchart of FIG. Thus, defective pixel correction can be performed on all of the image data before being divided. At this time, since any defective pixel address data can be registered in the storage unit 43 in any scene, the defective pixel correction unit 42 performs a single defective pixel correction operation to perform division according to the exposure time. it is possible to correct all the defective pixels of the divided image data I _SC of the divided image corresponding to the counter number C of the number S.

  In the defective pixel correction process of the second embodiment, the number of defective pixels to be corrected in the image sensor (CCD 20) exceeds the number of defective pixels (registration limit number) that can be registered in the storage unit 43 of the defective pixel correction unit 42. Even when the exposure time is 8 seconds, 16 seconds, or 32 seconds in the second embodiment, the registered content in the storage unit 43 is rewritten, that is, the registered content of defective pixel data in the storage unit 43 is changed. However, since the defective pixel correction operation in the defective pixel correction unit 42 is repeated, regardless of the relationship between the number of defective pixels that need to be corrected and the registration limit number in the storage unit 43, the imaging element (CCD 20) All defective pixels can be corrected.

In addition, the defective pixel correction processing according to the second embodiment is substantially the same as the case where the defective pixel correction unit 42 performs the defective pixel correction processing on the original image data regardless of the number of defective pixel correction operations in the defective pixel correction unit 42. It can be. This is the defective pixel correction unit 42 performs the defective pixel correction operation for each of the divided image data I _SC obtained by equally dividing the original image data in the dividing number S (step S25) since each defective pixel correction operation Since the data amount of 1 / S of the original image data has only to be processed in (Step S25), the processing time in each defective pixel correction operation (Step S25) is approximately 1 / S of the processing time for the original image data. Since the processing time in the defective pixel correction operation for the original image data is 1, the processing time required for correcting all the defective pixels in the original image data regardless of the division number S is as follows. dividing the image processing time of the data I _SC obtained by multiplying the number of divisions S can be expressed by ((1 / S) × number of divisions S = 1), due to the fact that can be made substantially equal to both . Therefore, regardless of the relationship between the number of defective pixels that need to be corrected and the registration limit number of the storage unit 43, all the defective pixels of the image sensor (CCD 20) can be corrected, and the defective pixel correcting unit 42 can be corrected. Thus, it is possible to suppress an increase in the processing time associated with repeating the defective pixel correction operation, that is, an increase in the processing time required for correcting all the defective pixels.

  Furthermore, in the defective pixel correction process according to the second embodiment, since the number of pixels to be corrected is changed according to the exposure time, it is possible to correct only the necessary number of correction pixels according to the exposure time. It is possible to prevent deterioration in image quality due to lack or overcorrection.

  In the defective pixel correction process according to the second embodiment, pixels having a high defective pixel level are preferentially corrected. Therefore, pixels that are easily recognized as defective pixels on an image (image data) can be preferentially corrected. The image quality can be improved more efficiently.

In the defective pixel correction process of the second embodiment, the maximum value of the division number S is set to 4, and the image data divided into four areas accordingly (divided images A, B, C, D (FIG. 8A) In addition, each priority correction pixel group P _NM is acquired and stored in the RAM 44, and the number of divisions S is set to 1, 2, and 4 (the possible values of the number of divisions S are about the maximum value). Therefore, it is possible to appropriately register the address data of each defective pixel that matches the number of corrected pixels according to the exposure time in the storage unit 43 of the defective pixel correcting unit 42, and to detect defective pixels. Data for each correction process (each priority correction pixel group P _NM ) can be efficiently stored in the RAM 44.

  For this reason, in the digital camera 1 of the imaging device according to the present invention, it is possible to appropriately correct all defective pixels to be corrected regardless of the registration limit number in the storage unit 43. As a result, even when dark current increases or accumulates, for example, an image (image data) taken in a situation where the temperature of the image sensor (CCD 20) rises or the exposure time is long, the quality is improved. Can be made.

  In particular, in the defective pixel correction process according to the second embodiment, all defective pixels of the image sensor (CCD 20) are corrected regardless of the relationship between the number of defective pixels that need to be corrected and the registration limit number in the storage unit 43. In addition, it is possible to suppress an increase in processing time as the defective pixel correction unit 42 repeats the defective pixel correction operation.

  In the above-described second embodiment, an example in which the exposure time is used as the determination criterion for the change in the influence of the defective pixel and the number of correction pixels according to the exposure time is set. As long as the correction pixel number is set on the basis of the factor that the influence of scratches changes, for example, the correction pixel number may be set based on the setting of the correction pixel number according to the ISO sensitivity. It is not limited to Example 2.

  In the second embodiment, the number of corrected pixels when the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42 is 500 and the setting of the number of corrected pixels according to the exposure time is an exposure time of 32 seconds. Although 2000 was set as the maximum, each numerical value is an example and is not limited thereto.

  Furthermore, in Example 2, the value that the division number S can take is set to a divisor of the maximum value. However, all defective pixels that need to be corrected are corrected by the defective pixel correction unit 42 using the defective pixel correction operation. In order to correct, the division number S is set according to the number of defective pixels (registration limit number) that can be registered in the storage unit 43, that is, the quotient obtained by dividing at least the correction pixel number by the registration limit number Any value may be used as long as a value obtained by rounding up the decimals is set as the division number S, and the present invention is not limited to the second embodiment. For example, in the second embodiment, when the exposure time is 16 seconds, the number of correction pixels is set to 1200, and the registration limit number of the storage unit 43 is 500. If the correction operation is performed three times, all 1200 defective pixels can be corrected. Therefore, the division number S may be set to 3. In this case, when storing the address data of each defective pixel in the RAM 44 in advance, it is necessary to store the address data of each defective pixel when the division number S is 3, that is, the image (image data) is divided into three. .

In the second embodiment, the division number S is set corresponding to the correction pixel number corresponding to the exposure time, and the preparation of defective pixel correction corresponding to the division number S and the exposure time is performed. Since each priority correction pixel group P _NM is acquired and stored in the RAM 44 for each image data of an area equally divided by the maximum value (4 in the second embodiment), the division number S is a divisor of the maximum value. (In Example 2, 1 or 2) (exposure time is 2, 4, and 8 seconds), the priority correction pixel groups P _NM for each image data equally divided by the maximum value are combined (step S32, step S34, step S37, and step S38), the number of divisions S may always be set to the maximum value to prepare for defective pixel correction.

In the second embodiment, each priority correction pixel group P _NM is acquired and stored in the RAM 44 for each image data of an area equally divided by the maximum value of the division number S (4 in the second embodiment). It is only necessary to be able to prepare for defective pixel correction according to the above, that is, the address data of each defective pixel can be appropriately registered in the storage unit 43 of the defective pixel correction unit 42, and is limited to the second embodiment. It is not a thing.

Next, an image pickup apparatus according to Embodiment 3 of the present invention and defective pixel correction processing performed therein will be described. The third embodiment is an example in which part of the processing content in the defective pixel correction process is different from the first and second embodiments. Since the basic configuration of the image pickup apparatus according to the third embodiment is the same as that of the digital camera 1 that is the image pickup apparatus according to the first embodiment described above, the same reference numerals are given to portions having the same configuration, and detailed description thereof will be given. Is omitted. FIG. 11 shows the number of defective pixel addresses (number of defective pixel data) stored in the RAM 44 in each priority correction pixel group P _N ′ or each priority correction pixel group P _NM ′. It is explanatory drawing similar to FIG. 4 and FIG.

  First, for the defective pixel correction process of the third embodiment, as in the first embodiment, the number of defective pixels appearing at each set exposure time is statistically grasped, and the defective pixels appearing according to the exposure time. The number that can be determined to be sufficient to correct all of the above is set as the number of corrected pixels according to the exposure time. In the third embodiment, the number of corrected pixels when the exposure time is 2 seconds is 200, the number of corrected pixels when the exposure time is 4 seconds is 500, and the number of corrected pixels when the exposure time is 8 seconds is 700. The number of corrected pixels when the exposure time is 16 seconds is 1000, and the number of corrected pixels when the exposure time is 32 seconds is 1500 (see FIG. 11). The number of corrected pixels corresponding to the exposure time is stored in the RAM 44.

  Here, the control unit 28 sets the maximum value of the number of corrected pixels according to the exposure time to 1500, and the number of defective pixels that can be registered in the storage unit 43 is 500. In order to correct all the subtracted defective pixels (1000) by the defective pixel correction unit 42, the maximum value of the division number S ′ is set to 2 based on the necessity of performing the defective pixel correction operation twice. . In the third embodiment, as described later, when correcting the number of pixels exceeding the number of defective pixels (registration limit number) that can be registered in the storage unit 43, first, the defective pixel correction unit 42 corrects the defective pixel of the registration limit number. Then, all of the remaining defective pixels are corrected for defective pixels. Therefore, the division number S ′ in the third embodiment is the same as that of the remaining defective pixels while suppressing an increase in processing time. In order to perform correction, in one piece of image data acquired by the CCD 20, the image data is equally divided in order to set a target region, that is, a data amount in one defective pixel correction operation in the defective pixel correction unit 42. Is a number.

  In the third embodiment, the control unit 28 converts the data of the detected defective pixel address (vertical and horizontal coordinates on the CCD 20) into each of the defective pixel correction processing with the number of corrected pixels corresponding to the set exposure time. In association with the defective pixel level and the set exposure time, it is stored in the RAM 44 as follows.

The basic defective pixel detection method in the third embodiment is the same as that in the first embodiment. However, as described above, first, the defective pixel correction unit 42 performs the correction of the defective pixel of the registration limit number. From all the pixels of the CCD 20, the registration limit number, that is, up to 500 pixels is detected in descending order of the defective pixel level, and 200 pixels from the top are used as address data of defective pixels in the first priority correction pixel group P ₁ ′. The remaining 300 pixels are stored in the RAM 44 as address data of defective pixels of the second priority correction pixel group P ₂ ′. The 200 defective pixels detected as the first priority correction pixel group P ₁ ′ become 200 defective pixels to be corrected when the exposure time is 2 seconds, and are detected as the second priority correction pixel group P ₂ ′. The 300 defective pixels thus added become 500 defective pixels to be corrected when the exposure time is 4 seconds by adding 200 defective pixels of the first priority correction pixel group P ₁ ′. In the third embodiment, the 500 defective pixels that are corrected when the exposure time is 4 seconds directly become defective pixels having a high correction priority.

Next, as described above, since the maximum value of the division number S ′ is set to 2, each image data divided into two regions (hereinafter referred to as divided images E and F (see FIG. 8B)). ), Each priority correction pixel group P _NM ′ is acquired. Here, N is the same as that in the second embodiment (however, since the first priority correction pixel group P ₁ ′ and the second priority correction pixel group P ₂ ′ have already been acquired, possible values are 3 to 5) and M in the third embodiment indicate which of the divided images E and F corresponds to each of the N-th group. It is fitted.

On the assumption that the defective pixel level is lower than the lowest defective pixel level in the second priority correction pixel group P ₂ ′ in each of the divided images E and F, the defective pixel level is 100 pixels in order from the highest pixel are detected as defective pixels, and are stored in the RAM 44 as address data of defective pixels in the third priority correction pixel group P _3M ′ (M = 1, 2). A combination of the 100 defective pixels detected as the third priority correction pixel group P ₃₁ ′ and the 100 defective pixels detected as the third priority correction pixel group P ₃₂ ′ (third priority) The correction pixel group P ₃ ′) is obtained by adding the 200 defective pixels of the first priority correction pixel group P ₁ ′ and the 300 defective pixels of the second priority correction pixel group P ₂ ′. When the exposure time is 8 seconds, 700 defective pixels are corrected.

Next, among the pixels of the divided images E and F, in the third priority correction pixel group P ₃ ′ (third priority correction pixel group P ₃₁ ′ or third priority correction pixel group P ₃₂ ′) in each pixel. Assuming that the defective pixel level is lower than that having the lowest defective pixel level, 150 pixels are detected as defective pixels in order from the pixel with the highest defective pixel level, and the fourth priority correction pixel group P _4M ′ (M = 1, 2) is stored in the RAM 44 as address data of defective pixels. A combination of 150 defective pixels detected as the fourth priority correction pixel group P ₄₁ ′ and 150 defective pixels detected as the fourth priority correction pixel group P ₄₂ ′ (fourth priority) The correction pixel group P ₄ ′) includes the 200 defective pixels of the first priority correction pixel group P ₁ ′, the 300 defective pixels and the third priority correction pixel of the second priority correction pixel group P ₂ ′. By adding the 200 defective pixels of the group P ₃ ′, 1000 defective pixels are corrected when the exposure time is 16 seconds.

Next, among the pixels of the divided images E and F, in the third priority correction pixel group P ₃ ′ (the fourth priority correction pixel group P ₄₁ ′ or the fourth priority correction pixel group P ₄₂ ′) in each pixel. Assuming that the defective pixel level is lower than the lowest defective pixel level, 250 pixels are detected as defective pixels in order from the pixel with the highest defective pixel level, and the fifth priority correction pixel group P _5M ′ (M = 1, 2) is stored in the RAM 44 as address data of defective pixels. A combination of 250 defective pixels detected as the fifth priority correction pixel group P ₅₁ ′ and 250 defective pixels detected as the fifth priority correction pixel group P ₅₂ ′ (fifth priority) The correction pixel group P ₅ ′) includes the 200 defective pixels of the first priority correction pixel group P ₁ ′, the 300 defective pixels of the second priority correction pixel group P ₂ ′, and the third priority correction pixels. By adding the 200 defective pixels of the group P ₃ ′ and the 200 defective pixels of the fourth priority correction pixel group P ₄ ′, 1500 defective pixels are corrected when the exposure time is 32 seconds.

In the control unit 28 of the third embodiment, as described above, address data (see FIG. 11) of each priority correction pixel group P _NM ′ (N = 1 to 5, M = 1, 2) stored in the RAM 44, that is, each defective pixel. A defective pixel correction process is performed using the data. FIG. 12 is a flowchart showing the defective pixel correction process contents of the third embodiment executed by the control unit 28, and FIG. 13 is a flowchart showing the process contents of step S56 in the flowchart of FIG. Hereinafter, each step of the flowchart of FIG. 12 (including each step of the flowchart of FIG. 13) will be described with reference to FIG.

  In step S51, the set exposure time is acquired, and the process proceeds to step S52. In Step S51, as in Step S1 of the flowchart of FIG. 5, when the release 2 is pushed down and the photographing operation is started, information on the exposure time at the time of photographing that can be obtained is acquired.

  In step S52, following the acquisition of the set exposure time in step S51, it is determined whether or not the acquired exposure time is 8 seconds or more. If Yes, the process proceeds to step S53. If No, step S60 is performed. Proceed to In step S52, the length of the exposure time acquired in step S51 is determined in order to switch the content of the defective pixel correction process, that is, to switch the number of defective pixel correction operations by the defective pixel correction unit 42. In the third embodiment, the number of defective pixels that can be registered (stored) in the storage unit 43 is 500, whereas the number of corrected pixels when the exposure time is 4 seconds is set to 500, and the exposure time is 8 seconds. In this case, since the number of corrected pixels is set to 700, the number of corrected pixels when the exposure time is 8 seconds or more exceeds the registration limit number. The process proceeds to step S53 so that the number of corrected pixels when the exposure time is 4 seconds or less is equal to or less than the registration limit number. Therefore, the defective pixel correction unit 42 performs the defective pixel correction operation once to proceed to step S60. move on.

In step S53, following the determination that the exposure time in step S52 is 8 seconds or longer, correction of defective pixels having a high correction priority is executed, and the flow proceeds to step S54. In step S53, since it is determined in step S52 that the exposure time is 8 seconds or more, it is necessary to correct more defective pixels than the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42. Therefore, the registration limit number of defective pixels in the storage unit 43 is corrected in descending order of correction priority, that is, in order of increasing defective pixel level (luminance value). In Example 3, since the registration limit number of the storage unit 43 is 500, the address data of 200 defective pixels of the first priority correction pixel group P ₁ ′ and 300 of the second priority correction pixel group P ₂ ′. The address data of the defective pixels is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42. Thereafter, the defective pixel correction unit 42 reads out pixel data corresponding to the address of the defective pixel registered in the storage unit 43 in the received image data (each pixel data), and looks at the surrounding RGB filters. By interpolating with pixel data corresponding to the same color, the total of 500 defective pixels of the first priority correction pixel group P ₁ ′ and 300 defective pixels of the second priority correction pixel group P ₂ ′ is 500. Correction of defective pixels is performed.

  In step S54, following the execution of correction of defective pixels having a high correction priority in step S53, the number of correction pixels corresponding to the exposure time acquired in step S51 is read, and the process proceeds to step S55. In step S54, the number of corrected pixels corresponding to the exposure time stored in the RAM 44 is read out.

In step S55, following the readout of the number of corrected pixels corresponding to the exposure time in step S54, a division number S ′ corresponding to the acquired exposure time is set, and the process proceeds to step S56. In step S55, a value obtained by subtracting the number of defective pixels (500) that can be registered in the storage unit 43 from the number of corrected pixels corresponding to the read exposure time is divided by the number of defective pixels (500), and the quotient is obtained. If the number is 1 or less, the division number S ′ is set to 1, and if the quotient is larger than 1 and 2 or less, the division number S ′ is set to 2. That is, of the number of corrected pixels corresponding to the exposure time, all the remaining corrected pixel numbers obtained by subtracting the number of defective pixels with high correction priority already corrected in step S53 are used by the defective pixel correction unit 42. The number of defective pixel correction operations required for correction is set as the division number S ′ (S = 1, 2). Specifically, the division number S ′ when the exposure time is 8 seconds is 1, the division number S ′ is 1 when the exposure time is 16 seconds, and the division number S ′ when the exposure time is 32 seconds. 2 (see FIG. 8B). Basically, the value obtained by subtracting the number of defective pixels (500) in the storage unit 43 from the number of corrected pixels corresponding to the read exposure time is divided by the number of defective pixels, and the decimal part of the quotient is rounded up. The obtained value is the division number S. In step S55, the counter number C is set to 1. This counter number C is the image data (divided image data I _SC ′ (S: division number, C: counter) of each divided image (E, F (see FIG. 8B)) equally divided by the division number S ′. It represents which of the number)), and 1 and 2 are applied in the order of E and F. Therefore, when the division number S'is 2, image data is divided into _'a divided image F of the divided image data I _22' divided image data I ₂₁ of the divided image E (FIG. 8 (b) see) .

In step S56, following the setting of the division number S ′ in step S55 or the increment of the counter number C in step S59, preparation for defective pixel correction is performed in accordance with the division number S ′ and the exposure time. Proceed to In this step S56, each of the priority correction pixel groups P _NM ′ (N = 3 to 3) is corrected in order to correct the pixels with the correction pixel number corresponding to the exposure time acquired in step S51 (substantially set the correction pixel number). 5, M = 1, 2) is read from the RAM 44 as appropriate and registered (stored) in the storage unit 43 of the defective pixel correction unit 42 to prepare for defective pixel correction according to the exposure time (this) Is treated as C). Below, each step of the flowchart of FIG. 13 which shows the processing content in the preparation (step S56) of the defective pixel correction | amendment according to this exposure time is demonstrated. As described above, when the process proceeds to the flowchart of FIG. 13, the process proceeds from step S51 → step S52 → step S53 → step S54 → step S55 in the flowchart of FIG. A total of 500 defective pixels having a high correction priority (200 defective pixels in the first priority correction pixel group P ₁ ′ and 300 defective pixels in the second priority correction pixel group P ₂ ′). (Defective pixel) has already been corrected.

  In step S71, it is determined whether or not the exposure time acquired in step S51 is 8 seconds. If Yes, the process proceeds to step S72. If No, the process proceeds to step S73.

In step S72, following the determination that the exposure time in step S71 is 8 seconds, preparation for defective pixel correction corresponding to the exposure time of 8 seconds is made, and the process C (step S56) is terminated, and step S57 is completed. Proceed to (see FIG. 12). Here, when the exposure time is 8 seconds, the division number S ′ is set to 1 in step S55 of the flowchart of FIG. In this step S72, as preparation for defective pixel correction corresponding to an exposure time of 8 seconds, the number of divisions is 1, so that all defective pixels in the divided image data I ₁₁ ′, which is image data as it is, are corrected for the first time. Address data of a total of 200 defective pixels of the address data (see FIG. 11) of each 100 defective pixels of the 3-priority correction pixel group P _3M ′ (M = 1, 2) is read from the RAM 44, and a defective pixel correction unit 42 is registered (stored) in the storage unit 43.

  In step S73, following the determination that the exposure time in step S71 is not 8 seconds, it is determined whether the exposure time acquired in step S51 is 16 seconds. If Yes, the process proceeds to step S74. In this case, the process proceeds to step S75.

In step S74, following the determination that the exposure time in step S73 is 16 seconds, preparation for defective pixel correction corresponding to the exposure time of 16 seconds is made, and the process C (step S56) is terminated, and step S57 is completed. Proceed to (see FIG. 9). Here, when the exposure time is 16 seconds, the division number S ′ is set to 1 in step S55 of the flowchart of FIG. In this step S74, as preparation for defective pixel correction according to the exposure time of 16 seconds, the number of divisions is 1, so that all the defective pixels in the divided image data I ₁₁ ′ as image data as they are are corrected. Address data of each 100 defective pixels of the 3 priority correction pixel group P _3M ′ (M = 1, 2) and 150 defective pixels of the 4th priority correction pixel group P _4M ′ (M = 1, 2). Address data (see FIG. 4) of a total of 500 defective pixels is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

  In step S75, following the determination that the exposure time in step S73 is not 16 seconds, it is determined whether or not the counter number C is 1, the process proceeds to step S76 if Yes, and in step S77 if No. Proceed to This means that if the exposure time is not 16 seconds in step S73, the upper limit of the set exposure time is set to 32 seconds. In step S55 of the flowchart of FIG. Therefore, as will be described later, the counter number C is increased from 1 to 2 in the flow of returning from step S56 to step S59 in the flowchart of FIG. 12 (step S59). ) Since the defective pixel correction unit 42 executes the defective pixel correction operation for each of the divided image E and the divided image F (see FIG. 8B), the defect corresponding to the exposure time of 32 seconds by the counter number C. This is due to different pixel correction preparations.

In step S76, following the determination that the counter number C in step S75 is 1, preparation for defective pixel correction of the divided image E (see FIG. 8B) corresponding to the exposure time of 32 seconds is performed. The process C (step S56) is terminated, and the process proceeds to step S57 (see FIG. 12). In this step S76, as preparation for defective pixel correction of the divided image E (see FIG. 8B) corresponding to the exposure time of 32 seconds, all defective pixels in the divided image data I ₂₁ ′ that is half of the image data are obtained. For correction, the Nth priority correction is a priority number (N = 3 to 5) corresponding to the exposure time of 32 seconds and a number (M = 1) corresponding to the divided image E obtained by dividing the image data into two. Address data (see FIG. 11) of each defective pixel of the pixel group P _NM ′ (N = 3 to 5, M = 1) is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42. That is, the address data of 100 defective pixels of the third priority correction pixel group P ₃₁ ′, the address data of 150 defective pixels of the fourth priority correction pixel group P ₄₁ ′, and the fifth priority correction pixel group P ₅₁ ′. A total of 500 defective pixel address data of 250 defective pixels (see FIG. 11) is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

In step S77, following the determination that the counter number C in step S75 is not 1, preparation for defective pixel correction of the divided image F (see FIG. 8B) corresponding to the exposure time of 32 seconds is performed. The process C (step S56) is terminated, and the process proceeds to step S57 (see FIG. 12). In this step S77, as preparation for defective pixel correction of the divided image F (see FIG. 8B) corresponding to the exposure time of 32 seconds, all defective pixels in the divided image data I ₂₂ ′ which is half of the image data are obtained. For correction, the Nth priority correction is a priority number (N = 3 to 5) corresponding to the exposure time of 32 seconds and a number (M = 2) corresponding to the divided image F obtained by dividing the image data into two. Address data (see FIG. 11) of each defective pixel of the pixel group P _NM ′ (N = 3 to 5, M = 2) is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42. That is, address data of 100 defective pixels of the third priority correction pixel group P ₃₂ ′, address data of 150 defective pixels of the fourth priority correction pixel group P ₄₂ ′, and the fifth priority correction pixel group P ₅₂ ′. A total of 500 defective pixel address data of 250 defective pixels (see FIG. 11) is read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

In step S57, following the preparation of defective pixel correction according to the number of divisions S ′ and the exposure time in step S56, the divided image data I of the divided image corresponding to the counter number C at the number of divisions S ′ according to the exposure time. Defective pixel correction is executed for _SC ′, and the process proceeds to step S58. In this step S57, the defective pixel correction unit 42 detects defective pixels registered in the storage unit 43 in the divided image data I _SC ′ corresponding to the counter number C in the transmitted image data (each pixel data). Pixel data corresponding to the address of the correction pixel corresponding to the exposure time in step S56 by interpolating with pixel data corresponding to the same color as seen by the surrounding RGB filters Correct the number.

At step S58, the continued execution of the defective pixel correction for the divided image data I _SC of the divided image corresponding to the counter number C in the divided number S'according to the exposure time in step S57, the count C division number S' If YES, the flowchart of FIG. 12 is terminated, and if NO, the process proceeds to step S59. In this step S58, it is determined whether or not the counter number C is equal to the division number S ′ corresponding to the exposure time, whereby step S57 for performing defective pixel correction on the divided image data I _SC ′ is repeated. Thus, it is determined whether or not defective pixel correction has been performed on all the divided images, that is, all of the image data before being divided.

In step S59, following the determination that the counter number C is not equal to the division number S ′ in step S58, the counter number C is incremented (C = C + 1), and the process returns to step S56. In this step S59, in order to perform defective pixel correction on all of the image data before division, the number of executions of defective pixel correction (counter number C) for the divided image data I _SC ′ in step S57 is incremented by one. to add.

In step S60, following the determination in step S52 that the exposure time is not 8 seconds or more, preparation for defective pixel correction according to the exposure time is performed, and the flow proceeds to step S61. In step S60, as in step S4 in the flowchart of FIG. 5, the address data of each priority correction pixel group P _N ′ is stored in the RAM 44 in order to correct the pixels having the number of correction pixels corresponding to the exposure time acquired in step S51. Are appropriately read out and registered (stored) in the storage unit 43 of the defective pixel correction unit 42 to prepare for defective pixel correction according to the exposure time. In this step S60, when the exposure time is 2 seconds, the address data (see FIG. 4) of 200 defective pixels of the first priority correction pixel group P ₁ ′ is prepared in the RAM 44 as preparation for defective pixel correction according to the exposure time. Are registered (stored) in the storage unit 43 of the defective pixel correction unit 42. When the exposure time is 4 seconds, the address data of 200 defective pixels of the first priority correction pixel group P ₁ ′ and the second priority correction pixel group P ₂ ′ are prepared as preparations for defective pixel correction according to the exposure time. Are read from the RAM 44 and registered (stored) in the storage unit 43 of the defective pixel correction unit 42.

In step S61, following the preparation for defective pixel correction corresponding to the exposure time in step S60, defective pixel correction corresponding to the exposure time is executed, and the flowchart of FIG. 12 ends. In step S61, the defective pixel correction unit 42 reads out pixel data corresponding to the address of the defective pixel registered in the storage unit 43 in the transmitted image data (each pixel data thereof), and the surrounding RGB filters In step S60, interpolation is performed using pixel data corresponding to the same color, thereby correcting the number of corrected pixels in accordance with the preparation for defective pixel correction in accordance with the exposure time in step S60. Specifically, when the exposure time is 2 seconds, 200 defective pixels of the first priority correction pixel group P ₁ ′ are corrected, and when the exposure time is 4 seconds, the first priority correction pixel group P ₁ ′ is corrected. A total of 500 defective pixels including 200 defective pixels and 300 defective pixels in the second priority correction pixel group P ₂ ′ are corrected. At this time, since all of the defective pixel address data can be registered in the storage unit 43 in any scene, the defective pixel correction unit 42 performs correction according to the exposure time by one defective pixel correction operation. All of the defective pixels of the number of pixels can be corrected. Therefore, in step S61, the exposure time is 2 seconds or 4 seconds, and the number of corrected pixels corresponding to the exposure time acquired in step S51 is used as the defective pixel correction operation in the first defective pixel correction unit 42. Correction is executed.

  As described above, in the defective pixel correction process according to the third embodiment, the number of defective pixels to be corrected in the image sensor (CCD 20) is within the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42. (In Example 3, when the exposure time is 2 seconds or 4 seconds), the process proceeds from step S51 to step S52 to step S60 to step S61. A defective pixel correction operation with the number of corrected pixels corresponding to the exposure time acquired by the pixel correction unit 42 is performed. For this reason, it is possible to perform defective pixel correction of the number of corrected pixels corresponding to the exposure time while making the processing content as a whole simpler.

When the number of defective pixels to be corrected in the image sensor (CCD 20) exceeds the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42 (in Example 3, the exposure time is 8 seconds and 16 seconds). Alternatively, in the case of 32 seconds), the process proceeds from step S51 to step S52 to step S53, and first, 500 defective pixels having a high correction priority (200 defective pixels in the first priority correction pixel group P ₁ ′ and the first) (300 defective pixels in the 2-priority correction pixel group P ₂ ′), and then proceeds to step S54 → step S55 → step S56 → step S57, and out of the number of correction pixels corresponding to the acquired exposure time The division number S ′ is set corresponding to the number of remaining defective pixels that have not been corrected (step S55), the image (image data) is equally divided by the division number S ′, and each divided image data is divided. By repeating the defective pixel correction operation in the defective pixel correction unit 42 with respect to data I _{SC ',} performs defective pixel correction for all of the image data without overlapping the pixel data to be corrected (step S57). At this time, in step S56, the address data of each defective pixel in the divided image data I _SC ′ corresponding to the number of executions of the defective pixel correction (counter number C) in step S57 is stored in the RAM 44. By repeating the flow from S56 → step S57 → step S58 → step S59 and returning to step S56, defective pixel correction can be performed on all of the image data before being divided. For example, when the exposure time is set to 32 seconds (the number of correction pixels is 1500), the division number S ′ is set by proceeding from step S51 → step S52 → step S53 → step S54 → step S55. 2 and the image data is equally divided into divided images E and F. Thereafter, preparation for defective pixel correction of the divided image data I ₂₁ ′ of the divided image E corresponding to the exposure time of 32 seconds is performed in step S56, and the process proceeds to step S57, whereby the defective pixel correction unit 42 is sent. In the divided image data I ₂₁ ′ of the received image data, the pixel data corresponding to the address of the defective pixel registered in the storage unit 43 (100 defective pixels of the third priority correction pixel group P ₃₁ ′, fourth 150 defective pixels in the priority correction pixel group P ₄₁ ′ and 250 defective pixels in the fifth priority correction pixel group P ₅₁ ′ (total 500 defective pixels) are read out and correspond to the same color as seen by the surrounding RGB filters The defective pixel correction corresponding to the exposure time of 32 seconds of the divided image E is completed by interpolating with the pixel data. Thereafter, the process proceeds to step S58, but since the counter number C is 1 and is not equal to the division number S ′ = 2, the process proceeds to step S59 and returns to step S56, thereby dividing the exposure time according to 32 seconds. Preparation for defect pixel correction of the divided image data I ₂₂ ′ of the image F is performed. Thereafter, by proceeding to step S57, pixel data corresponding to the address of the defective pixel registered in the storage unit 43 in the divided image data I ₂₂ ′ of the image data sent to the defective pixel correction unit 42. (100 defective pixels in the third priority correction pixel group P ₃₂ ′, 150 defective pixels in the fourth priority correction pixel group P ₄₂ ′, and 250 defective pixels in the fifth priority correction pixel group P ₅₂ ′ A total of 500 defective pixels) is read out and interpolated with pixel data corresponding to the same color as seen by the surrounding RGB filters, and the defective pixel correction of the divided image F is completed. Thereafter, when the process proceeds to step S58, since the counter number C is 2, the process does not proceed to step S59, and the flowchart of FIG. After the correction is completed, defective pixel correction can be performed on all of the image data before being divided. At this time, since any defective pixel address data can be registered in the storage unit 43 in any scene, the defective pixel correction unit 42 performs a single defective pixel correction operation to perform division according to the exposure time. All of the defective pixels in the divided image data I _SC ′ of the divided image corresponding to the counter number C in the number S ′ can be corrected.

  In the defective pixel correction process of the third embodiment, when the number of defective pixels to be corrected in the image sensor (CCD 20) exceeds the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42 (in the third embodiment). Even when the exposure time is 8 seconds, 16 seconds, or 32 seconds, the defective pixel correction is performed while rewriting the registration content in the storage unit 43, that is, changing the registration content of the defective pixel data in the storage unit 43. Since the defective pixel correction operation in the unit 42 is repeated, all defective pixels in the image sensor (CCD 20) are corrected regardless of the relationship between the number of defective pixels that need correction and the registration limit number in the storage unit 43. can do.

  In the defective pixel correction process of the third embodiment, when the number of defective pixels to be corrected in the image sensor (CCD 20) does not exceed the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42, The defective pixel correction can be performed more appropriately in the processing time of one defective pixel correction operation in the defective pixel correction unit 42 for the entire image (image data). This is because the defective pixel correction unit 42 performs the defective pixel correction operation on the entire image (image data), so that the probability that there is a pixel adjacent to the defective pixel of interest increases, and the entire image (image data). Even if there is a bias in the distribution of each defective pixel in the above, it is possible to reliably preferentially correct a pixel having a high defective pixel level regardless of the bias.

Furthermore, in the defective pixel correction process of the third embodiment, the number of defective pixels to be corrected in the image sensor (CCD 20) exceeds the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correction unit 42. In addition, the defective pixel correction can be performed more appropriately in the processing time of the defective pixel correction operation for approximately two times in the defective pixel correction unit 42 for the entire image (image data). That is, for the entire image (image data), first, 500 defective pixels having a high correction priority are corrected, and then the division number S ′ is set for each remaining defective pixel. Since the defective pixel correction operation is performed on each divided image data I _SC ′ obtained by equally dividing the image data by the division number S ′ (step S57), the defective pixel correction is performed on the entire image (image data). While harmonizing both the effect that a more appropriate correction can be performed by the above and the effect of suppressing an increase in processing time by correcting the defective pixel by equally dividing the original image data by the division number S ′ Can be obtained.

  In the defective pixel correction process according to the third embodiment, since the number of pixels to be corrected is changed according to the exposure time, it is possible to correct only the necessary number of correction pixels according to the exposure time. It is possible to prevent a deterioration in image quality due to overcorrection.

  In the defective pixel correction process according to the third embodiment, since pixels with high defective pixel levels are preferentially corrected, pixels that are easily recognized as defective pixels on an image (image data) can be preferentially corrected. The image quality can be improved more efficiently.

  For this reason, in the digital camera 1 of the imaging device according to the present invention, it is possible to appropriately correct all defective pixels to be corrected regardless of the registration limit number in the storage unit 43. As a result, even when dark current increases or accumulates, for example, an image (image data) taken in a situation where the temperature of the image sensor (CCD 20) rises or the exposure time is long, the quality is improved. Can be made.

  In particular, in the defective pixel correction process according to the third embodiment, all defective pixels of the image sensor (CCD 20) are more appropriately determined regardless of the relationship between the number of defective pixels that need correction and the registration limit number of the storage unit 43. In addition to correction, it is possible to suppress an increase in processing time due to the defective pixel correction unit 42 repeating the defective pixel correction operation.

  In the above-described third embodiment, an example is shown in which the number of correction pixels corresponding to the exposure time is set with the exposure time as the criterion for determining the change in the influence of defective pixels. For example, the correction pixel number may be set based on the setting of the correction pixel number according to the ISO sensitivity. It is not limited to three.

  Further, in the third embodiment, the number of corrected pixels when the number of defective pixels that can be registered in the storage unit 43 of the defective pixel correcting unit 42 is 500 and the setting of the number of corrected pixels according to the exposure time is an exposure time of 32 seconds. Although 1500 is set as the maximum, each numerical value is an example and is not limited thereto. Here, when the set number of correction pixels is three times or more of the registration limit number, at least a quotient obtained by dividing the registration pixel number by the registration limit number by subtracting the registration limit number from the correction pixel number. What is necessary is just to set the value which rounded up the lesser than this as division | segmentation number S '.

Furthermore, in the third embodiment, the division number S ′ is set corresponding to the number of correction pixels corresponding to the exposure time, and preparation for defective pixel correction corresponding to the division number S ′ and the exposure time is performed. Since each priority correction pixel group P _NM ′ is acquired and stored in the RAM 44 for each image data of an area equally divided by the maximum value of the division number S ′ (2 in the third embodiment), the division number S When 'is a divisor of the maximum value (1 in the third embodiment) (exposure time is 8 and 16 seconds), each priority correction pixel group P _NM ' for each image data equally divided by the maximum value is combined. Therefore (see step S72 and step S74), the number of divisions S ′ may always be set to the maximum value to prepare for defective pixel correction.

In the third embodiment, each priority correction pixel group P _NM ′ is acquired and stored in the RAM 44 for each image data of an area equally divided by the maximum value of the division number S ′ (2 in the third embodiment). Any device may be used as long as it can prepare for defective pixel correction according to the exposure time, that is, can register address data of each defective pixel in the storage unit 43 of the defective pixel correction unit 42 as appropriate. Is not to be done.

  As described above, the present invention has been described in detail based on the respective embodiments. However, the present invention is not limited to these specific configurations, and design changes that do not depart from the spirit of the present invention are included in the technical scope of the present invention.

  In each of the above-described embodiments, the number of corrected pixels corresponding to each exposure time is set, and defective pixels at each exposure time are detected based on the correction pixel number. However, in actuality, a digital still camera (digital camera 1) is used. ) In the state where the sensor (CCD 20) is mounted, the luminance value of the pixel (pixel data) appearing as a bright spot in the acquired dark image (image data) at each set exposure time is grasped, and according to the exposure time. A luminance value that can be determined to be sufficient for correcting all of the appearing bright spots (pixels) is set as a threshold value (correction determination value) of a defective pixel level corresponding to the exposure time, and the defective pixel corresponding to the exposure time is set. A defective pixel at each exposure time may be detected using the level as a criterion. This is the same even when the ISO sensitivity is used as a criterion for determining the change in the influence of defective pixels (temperature scratches).

  In each of the above-described embodiments, the RGB filter is provided in the image sensor (CCD 20). However, the image sensor (CCD 20) may be configured without the filter. In this case, in the defective pixel correction in the defective pixel correction unit 42, the average value of the pixel data on both sides adjacent to the focused defective pixel is set as the pixel data of the focused defective pixel without considering the color classification. Interpolation may be performed by

DESCRIPTION OF SYMBOLS 1 (As imaging device) Digital camera 21 Analog front end part (As image generation means) 22 Signal processing part (As image generation means) 23 SDRAM (As image generation means)
24 ROM (as image generation means)
28 Control unit (as image generation means) 42 Defective pixel correction unit 43 Storage unit 44 RAM (as memory)
I _SC divided image data P _N , P _NM (as defective pixel data) priority correction pixel group S, S ′ number of divisions

JP 2002-94884 A

Claims (11)

Image generating means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels;
When the image generation data in the image data generation process is input, the image generation unit corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. An imaging apparatus having a defective pixel correction unit that performs defective pixel correction of the image generation data by a defective pixel correction operation,
The image generation unit sets the number of correction pixels to be corrected based on a defective pixel level associated with a criterion for determining a change in the influence of a defective pixel, and the correction pixel number can be registered in the storage unit. When the registration limit number that is the number of defective pixel data is exceeded, the registration content of the defective pixel data in the storage unit is determined so that defective pixel correction of the correction pixel number is performed without duplicating pixel data to be corrected. While repeating the defective pixel correction operation in the defective pixel correction unit, the image generation data is a value obtained by rounding up the decimal of the quotient obtained by dividing the number of corrected pixels by the registration limit number Dividing equally into a plurality of divided image data, and changing the registered content of the defective pixel data in the storage unit for each divided image data, the defective pixel in the defective pixel correction unit Imaging apparatus characterized by repeating the positive operation. Image generating means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels;
When the image generation data in the image data generation process is input, the image generation unit corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. An imaging apparatus having a defective pixel correction unit that performs defective pixel correction of the image generation data by a defective pixel correction operation,
The image generation unit sets the number of correction pixels to be corrected based on a defective pixel level associated with a criterion for determining a change in the influence of a defective pixel, and the correction pixel number can be registered in the storage unit. When the registration limit number that is the number of defective pixel data is exceeded, the registration content of the defective pixel data in the storage unit is determined so that defective pixel correction of the correction pixel number is performed without duplicating pixel data to be corrected. The defective pixel correction operation in the defective pixel correction unit is repeated while changing, and the defective pixel data of the registration limit number is registered in the storage unit for the image generation data, and the defective pixel correction unit The defective pixel correction operation is performed, and then the image generation data is equal to or smaller than the quotient obtained by dividing the difference obtained by subtracting the registration limit number from the correction pixel number by the registration limit number. Divided into a plurality of divided image data by equally dividing the rounded-up value, and for each of the divided image data, the pixel data to be corrected does not overlap with the defective pixel correction in the first defective pixel correction operation. An imaging apparatus , wherein the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registered content of the defective pixel data in a storage unit . Image generating means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels;
When the image generation data in the image data generation process is input, the image generation unit corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. An imaging apparatus having a defective pixel correction unit that performs defective pixel correction of the image generation data by a defective pixel correction operation,
The image generation unit sets the number of correction pixels to be corrected based on a defective pixel level associated with a criterion for determining a change in the influence of a defective pixel, and the correction pixel number can be registered in the storage unit. When the registration limit number that is the number of defective pixel data is exceeded, the registration content of the defective pixel data in the storage unit is determined so that defective pixel correction of the correction pixel number is performed without duplicating pixel data to be corrected. The defective pixel correction operation in the defective pixel correction unit is repeated while changing, and the defective pixel data of the registration limit number is sequentially stored in the storage unit with respect to the image generation data in descending order of the defective pixel level. register performs the defective pixel correction operation in the defective pixel correction unit, then the image generation data, wherein said correction difference number of pixels obtained by subtracting the number of the registration limit registration limit Pixel data is equally divided by a value obtained by rounding up a few following quotient obtained by dividing the number is divided into a plurality of divided image data, and defective pixel correction of the first of the defective pixel correction operations to be corrected the defective pixel correction operation characterized imaging device to repeat the but in the defective pixel correction unit while changing the registration contents of the defective pixel data in the storage unit for each of the divided image data so as not to overlap . The imaging apparatus according to any one of claims 1 to 3, wherein
Furthermore, the image generating means comprises a memory that can be stored and read as appropriate.
The defective pixel data is determined based on an electrical signal output from the image sensor, and the defective pixel level of the pixel that has exceeded the correction determination value associated with the setting condition in the determination criterion. The ordinate and abscissa on the image sensor are generated in association with the corresponding setting condition and the defective pixel level and stored in the memory,
It said image generating means, you characterized imaging device to register in the storage unit of the defective pixel data conforms to the setting conditions are read from the memory. The defective pixel data is determined by determining the defective pixel level based on an electrical signal output from the image sensor for each section on the image sensor corresponding to the divided image data that has been equally divided. The ordinate and the horizontal coordinate on the image sensor of the pixel whose level exceeds the correction judgment value associated with the setting condition in the judgment criterion are generated by associating the corresponding setting condition and the defective pixel level and storing them in the memory the imaging apparatus according to any one of claims 1 to 4, characterized in that it is. The criterion is the exposure time, the setting condition is imaging apparatus according to any one of claims 1 to 5, characterized in numerical der Rukoto set as the exposure time. The criterion is the ISO sensitivity, the setting condition imaging apparatus according to any one of claims 1 to claim 5, characterized in that a value set as the ISO sensitivity. Image generating means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels;
When the image generation data in the image data generation process is input, the image generation unit corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. A program for causing a control unit of an imaging apparatus having a defective pixel correction unit that performs defective pixel correction of the image generation data to realize a function of correcting a defective pixel by the defective pixel correction unit by a defective pixel correction operation. ,
A function of causing the image generation means to set the number of correction pixels to be corrected based on a defective pixel level associated with a determination criterion for a change in the influence of a defective pixel;
When the number of corrected pixels exceeds the registration limit number that is the number of defective pixel data that can be registered in the storage unit, defective pixel correction is performed for the number of corrected pixels without overlapping pixel data to be corrected. Therefore, the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registration content of the defective pixel data in the storage unit, and the number of corrected pixels is set to the registration limit number. Dividing into a plurality of divided image data by equally dividing a fractional number obtained by dividing by a rounded up value, and changing the registration content of the defective pixel data in the storage unit for each divided image data A function of repeating the defective pixel correction operation in the defective pixel correction unit;
Program characterized Rukoto is realized. Image generating means for generating image data based on a plurality of electrical signals output from an imaging device having a plurality of pixels;
When the image generation data in the image data generation process is input, the image generation unit corrects pixel data corresponding to each defective pixel data in the image sensor registered in the storage unit in the image generation data. A program for causing a control unit of an imaging apparatus having a defective pixel correction unit that performs defective pixel correction of the image generation data to realize a function of correcting a defective pixel by the defective pixel correction unit by a defective pixel correction operation. ,
A function of causing the image generation means to set the number of correction pixels to be corrected based on a defective pixel level associated with a criterion for determining a change in the influence of a defective pixel;
When the number of corrected pixels exceeds the registration limit number that is the number of defective pixel data that can be registered in the storage unit, defective pixel correction is performed for the number of corrected pixels without overlapping pixel data to be corrected. Therefore, the defective pixel correction operation in the defective pixel correction unit is repeated while changing the registration content of the defective pixel data in the storage unit , and the registration limit number of the defective pixels is changed with respect to the image generation data. Data is registered in the storage unit to cause the defective pixel correction unit to perform the defective pixel correction operation, and then the image generation data is determined by subtracting the registration limit number from the correction pixel number. Dividing into a plurality of divided image data by equally dividing the value obtained by dividing the quotient obtained by dividing by the registration limit number into a plurality of divided image data, the defective pixel correction in the first defective pixel correction operation is a correction pair. A function to repeat the defective pixel correction operation in the defective pixel correction unit while changing the registration contents of the defective pixel data in the storage unit so that the pixel data do not overlap each said divided image data to,
A program characterized by realizing. The defective pixel level is determined based on an electrical signal output from the image sensor to the image generation unit, and the defective pixel level of the pixel that exceeds the correction determination value associated with the setting condition in the determination criterion Among the defective pixel data generated by associating the corresponding setting condition and the defective pixel level with the vertical and horizontal coordinates on the image sensor and stored in the memory, the defective pixel data that matches the setting condition is The program according to claim 8 or 9, wherein a function of reading from a memory and registering in the storage unit is realized. The defective pixel level is determined on the basis of an electrical signal output from the image sensor for each section on the image sensor corresponding to the divided image data equally divided as the defective pixel data, and the defective pixel The ordinate and the horizontal coordinate on the image sensor of the pixel whose level exceeds the correction judgment value associated with the setting condition in the judgment criterion are generated by associating the corresponding setting condition and the defective pixel level and storing them in the memory the ability to program according to any one of claims 8 to claim 10, characterized in that to realize.

Images