JP2004159087A - Scanner system and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a scanner system capable of satisfactorily reducing the granular noise without depending on the luminance change. <P>SOLUTION: This scanner system comprises a CCD 4 for imaging a film 2 already developed and loading a video signal of the image, a luminance detecting part 8 for detecting the luminance information at each local area with a prescribed size from the video signal loaded by the CCD 4, film noise estimating part 9 for estimating the granular noise quantity of the film 2 at each local area based on the luminance information detected by the detecting part 8, and a noise reducing part 10 for reducing the noise of the video signal at each local area based on the granularity noise quantity estimated by the estimating part 9. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a scanner system and an image processing program capable of reducing at least one of granularity noise caused by a film and element noise caused by an image sensor.
[0002]
[Prior art]
2. Description of the Related Art A variety of scanner systems have been proposed in the related art, in which an image on a film is picked up by an image pickup device and image data to be output to a device such as a personal computer or a printer is generated.
[0003]
Such a scanner system (film scanner system) is often configured to perform various types of image processing on a captured image. As an example of the image processing, a process of removing granularity noise specific to a film is performed. No. This granularity noise is a noise in which a fine density variation spatially occurs due to the density of the pigment cloud, and is visually observed as a gray-scale granular pattern.
[0004]
As a specific technique for removing such granularity noise, for example, Japanese Patent Application Laid-Open No. 9-51444 (Patent Document 1) discloses an image based on density information of a color film to be read and film sensitivity. A contour correction method for a color scanner that determines noise removal characteristics in a contour component signal extracted from data is described.
[0005]
Japanese Patent Application Laid-Open No. 2002-229136 (Patent Document 2) discloses that a film after development processing is detected by light in the infrared wavelength region (infrared light), and the detected information remains in a processed image. An image forming method for estimating a developed silver amount and performing a color correction process on digital image information based on a preset relational expression between the developed silver amount and the color correction processing amount is described.
[0006]
By the way, it is known that the granularity changes according to the brightness (density) of an image. For example, “Revision of Basics of Photographic Optics, Silver Halide Photography” (Corona) (Non-Patent Document 1), page 506 (See FIG. 14) showing the relationship between luminance (density) and granularity of a typical color film. FIG. 14 is a diagram showing a typical example of how the RMS granularity changes with respect to the film density.
[0007]
FIG. 14A shows the RMS granularity of a typical color negative film (ISO 100). After having a peak in a relatively low density range, the RMS granularity σA (D) increases as the density D increases. It is slowly decreasing.
[0008]
FIG. 14B shows the RMS granularity of a typical color reversal film (ISO 100). As the density D increases, the RMS granularity σA (D) monotonically increases.
[0009]
Such a granularity noise is determined by a dye cloud in the case of a color film, and the dye cloud is formed of silver halide grains as described in detail in p. Grown at the time of development centering on the development starting point, and having a diameter of about 3 μm to 20 μm. When the luminance at the time of exposure increases, the degree of growth of the silver halide increases, and the size of the dye cloud also tends to increase.
[0010]
By adjusting the type of silver halide particles, the type of coupler, and the coating density in consideration of the characteristics of the granularity noise caused by the dye cloud, the film has a high granularity that is hardly visually conspicuous. The graininess is designed to be better in the low density range (white in the positive state) where the graininess is more noticeable than in the density range (black in the positive state) (see FIG. 14 above). As the magnification of the film is increased, the granularity noise is visually recognized.
[0011]
On page 539 of Non-Patent Document 1, it is described that the pixel size, which is the minimum unit of spatial recording on a film, changes depending on the difference in ISO sensitivity, and is 12 μm in ISO 100 and 12 μm in ISO 200. It is 14 μm, 19 μm for ISO 400, 26 μm for ISO 1000, and the like. This indicates that the pixel size increases as the ISO sensitivity increases.
[0012]
Further, in a scanner system, it is known that random noise having white noise characteristics is generated as element noise from an imaging system including an imaging element. As a technique to cope with this element noise, for example, Japanese Patent Application Laid-Open No. 2001-157057 (Patent Document 3) describes a means for estimating a noise amount with respect to a signal level and controlling a frequency characteristic of filtering.
[0013]
[Patent Document 1]
JP-A-9-51444
[0014]
[Patent Document 2]
JP-A-2002-229136
[0015]
[Patent Document 3]
JP 2001-157057 A
[0016]
[Non-patent document 1]
"Revised basics of photo optics, silver halide photography" (Corporation) Photographic Society of Japan (Corona)
[0017]
[Problems to be solved by the invention]
However, since the technology described in Patent Documents 1 and 2 is a technology for uniformly removing noise due to individual differences between films, it corresponds to a case where granularity noise changes depending on the luminance level of an image. I can't.
[0018]
Further, the technique disclosed in Patent Document 3 does not consider reducing the element noise and the granularity noise at the same time.
[0019]
The present invention has been made in view of the above circumstances, and provides a scanner system capable of obtaining a high-quality image in which at least one of granularity noise and element noise is favorably reduced without depending on a change in luminance. It is intended to provide programs.
[0020]
[Means for Solving the Problems]
In order to achieve the above object, a scanner system according to a first aspect of the present invention is a scanner system that captures a developed film by an image sensor and captures a video signal related to an image on the film. Brightness detecting means for detecting brightness information for each area of size; film noise estimating means for estimating the granularity noise amount of the film for each area based on the brightness information detected by the brightness detecting means; Noise reduction means for reducing the noise of the video signal based on the granularity noise amount estimated by the estimation means.
[0021]
A scanner system according to a second aspect of the present invention is a scanner system that captures a developed film by an image sensor and captures a video signal related to an image on the film. , A film noise estimating means for estimating a granularity noise amount of the film for each of the regions based on the luminance information detected by the luminance detecting means, and a luminance information detected by the luminance detecting means. Element noise estimating means for estimating the element noise amount of the image sensor for each area based on the above, and the granularity noise amount estimated by the film noise estimating means and the element noise amount estimated by the element noise estimating means. Noise reduction means for reducing the noise of the video signal based on the noise.
[0022]
The scanner system according to a third aspect of the present invention is the scanner system according to the first or second aspect, wherein the film noise estimating means includes an operating means for selecting the type of the film, and a plurality of types of film granularity. Granularity noise recording means for recording the granularity noise characteristic, and selecting means for selecting the granularity noise characteristic recorded in the granularity noise recording means based on the type of film selected by the operating means. It is configured to estimate the granularity noise amount of the film from the luminance information based on the granularity noise characteristic selected by the selecting means.
[0023]
A scanner system according to a fourth aspect of the present invention is the scanner system according to the first or second aspect, wherein the film noise estimating means includes: an extracting means for extracting a video signal of an unexposed area from the video signal; Calculating means for calculating film characteristics based on the extracted video signal of the unexposed area, granularity noise recording means for recording granularity noise characteristics of a plurality of types of films, and a film calculated by the calculating means Selecting means for selecting the granularity noise characteristic recorded in the granularity noise recording means on the basis of the characteristic, and selecting the luminance information based on the granularity noise characteristic selected by the selecting means. Is used to estimate the granularity noise amount of the film.
[0024]
A scanner system according to a fifth aspect of the present invention is the scanner system according to the third or fourth aspect, wherein the film noise estimating means further includes an applying means for applying a predetermined granularity noise characteristic, and When it is not possible to select the granularity noise characteristic recorded in the granularity noise recording means by the means, the granularity of the film is determined from the luminance information based on the predetermined granularity noise characteristic provided by the applying means. This is to estimate the amount of noise.
[0025]
A scanner system according to a sixth aspect of the present invention is the scanner system according to the fourth aspect, wherein the calculating means calculates a variance of the video signal in the unexposed area and calculates a film characteristic based on the variance. It is.
[0026]
A scanner system according to a seventh aspect of the present invention is the scanner system according to the second aspect, wherein the element noise estimating means includes: an element noise recording means for recording element noise characteristics of the image sensor; Gain correction means for reading the element noise characteristics and correcting based on the gain for the video signal, and for each of the regions based on the element noise characteristics corrected by the gain correction means and the luminance information. This is for estimating the element noise amount of the imaging element.
[0027]
The scanner system according to an eighth aspect is the scanner system according to the first aspect, wherein the noise reduction unit is set by a filter setting unit that sets a filter characteristic based on the granularity noise amount, and the filter setting unit. Filtering means for filtering the video signal using the filter characteristics to reduce noise.
[0028]
A scanner system according to a ninth aspect of the present invention is the scanner system according to the second aspect, wherein the noise reduction unit adds the granularity noise amount and the element noise amount by the addition unit and the addition unit. Filter setting means for setting filter characteristics based on the amount of noise, and filtering means for reducing noise by filtering the video signal using the filter characteristics set by the filter setting means It was done.
[0029]
A scanner system according to a tenth aspect of the present invention is the scanner system according to the first aspect, wherein the noise reduction unit sets an upper limit value and a lower limit value based on the granularity noise amount and the luminance information. And a correction unit that corrects the video signal based on the upper limit value and the lower limit value set by the allowable range setting unit to reduce noise.
[0030]
In a scanner system according to an eleventh aspect of the present invention, in the scanner system according to the second aspect, the noise reduction unit is configured to add the granularity noise amount and the element noise amount by the addition unit and the addition unit. An allowable range setting means for setting an upper limit value and a lower limit value based on the noise amount and the luminance information; and a noise correcting device for correcting the video signal based on the upper limit value and the lower limit value set by the allowable range setting means. And correction means for reducing
[0031]
A scanner system according to a twelfth aspect, in the scanner system according to the first aspect, wherein the noise reduction unit sets a noise amplitude value as a threshold based on the granularity noise amount, and the threshold setting. And a smoothing means for reducing an amplitude component of the video signal which is equal to or less than a threshold value set by the means.
[0032]
A scanner system according to a thirteenth aspect of the present invention is the scanner system according to the second aspect, wherein the noise reduction unit is configured to add the granularity noise amount and the element noise amount by the addition unit and the addition unit. Threshold value setting means for setting the amplitude value of the noise as a threshold value based on the noise amount; and smoothing means for reducing the amplitude component of the video signal which is equal to or less than the threshold value set by the threshold value setting means. It is a thing.
[0033]
An image processing program according to a fourteenth aspect of the present invention provides a computer-readable storage medium storing a computer program for controlling a computer to detect luminance information for each area of a predetermined size from a video signal of an image on a film captured by capturing an image of the developed film by an image sensor. Means, a film noise estimating means for estimating a granularity noise amount of the film for each area based on the luminance information detected by the luminance detecting means, and a film noise estimating means for estimating the granularity noise amount estimated by the film noise estimating means. This is a program that functions as a noise reduction unit that reduces noise of a video signal.
[0034]
An image processing program according to a fifteenth aspect of the present invention provides a computer-readable storage medium that controls a computer to detect luminance information for each region of a predetermined size from a video signal of an image on a film captured by capturing an image of a developed film by an image sensor. Means, a film noise estimating means for estimating the granularity noise amount of the film for each area based on the luminance information detected by the luminance detecting means, and a film noise estimating means for each area based on the luminance information detected by the luminance detecting means. An element noise estimating means for estimating an element noise amount of the image sensor; a noise of the video signal based on the granularity noise amount estimated by the film noise estimating means and the element noise amount estimated by the element noise estimating means. This is a program that functions as noise reduction means for reducing noise.
[0035]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
1 to 6 show a first embodiment of the present invention, and FIG. 1 is a block diagram showing a configuration of a scanner system.
[0036]
The scanner system includes a light source 1, a lens system 3 for imaging transmitted light from a film 2 illuminated by the light source 1, and an optical image formed by the lens system 3 by photoelectric conversion and analog conversion. CCD 4 serving as an image pickup device for outputting a video signal as an image signal, an amplifier 5 for amplifying a video signal output from the CCD 4, and an A / D converter for converting an analog video signal amplified by the amplifier 5 into a digital video signal. Device 6, an image buffer 7 for temporarily storing the video signal converted by the A / D converter 6, and an area of a predetermined size (for example, 2 × 2) from the video signal stored in the image buffer 7. (A pixel-based area) to extract an average luminance value and calculate the average luminance value. The luminance detection unit 8 and a video signal of an unexposed area in the video signal stored in the image buffer 7 A film noise estimating unit 9 serving as a film noise estimating unit for estimating a granularity noise characteristic and estimating a granularity noise amount based on the granularity noise characteristic and the average luminance value calculated by the luminance detecting unit 8; A noise reduction unit 10 serving as a noise reduction unit for reading a video signal stored in the buffer 7 for each region of a predetermined size and performing a noise reduction process based on the granularity noise amount estimated by the film noise estimation unit 9; An image processing unit 11 that performs known enhancement processing, compression processing, and the like on the video signal whose noise has been reduced by the unit 10, and records and saves the video signal processed by the image processing unit 11 on, for example, a disk-shaped recording medium. An output unit 12, an external I / F unit 14 serving as operating means having various interfaces, the amplifier 5, the A / D converter 6, and the luminance The scanner system is bidirectionally connected to the detecting unit 8, the film noise estimating unit 9, the noise reducing unit 10, the image processing unit 11, the output unit 12, and the external I / F unit 14, and integrally controls the scanner system including these components. And a control unit 13 serving as a control unit also serving as an operation unit including a microcomputer or the like.
[0037]
Examples of the interface provided in the external I / F unit 14 include a power switch for turning on / off the power of the scanner system, a shutter button for instructing the CCD 4 to perform imaging, A mode switching button for switching the mode, a film type selection operation unit for specifying the type of the film 2 and the like are included.
[0038]
It should be noted that the film 2 is to be loaded after being appropriately replaced by a user, and is not provided uniquely to this scanner system. The film 2 loaded in the scanner system has already been developed as described later with reference to FIG.
[0039]
The operation of such a scanner system will be described along a signal flow.
[0040]
When the light source 1 emits light, the illumination light is applied to the film 2 to be scanned, and the light transmitted through the film 2 is formed as an optical image on the imaging surface of the CCD 4 by the lens system 3.
[0041]
The CCD 4 converts the optical image into a video signal and outputs the video signal. The output video signal is amplified by the amplifier 5 by a predetermined amplification amount.
[0042]
The amplified video signal is stored in the image buffer 7 after being converted into a digital signal by the A / D converter 6.
[0043]
Under the control of the control unit 13, the luminance detection unit 8 extracts an area of a predetermined size, for example, an area of 2 × 2 pixels, from the video signal stored in the image buffer 7 and calculates an average luminance value. Then, the data is transferred to the film noise estimating unit 9.
[0044]
The film noise estimating unit 9 extracts a video signal of an unexposed area (see FIG. 3) from the video signal stored in the image buffer 7 based on the control of the control unit 13, and performs dispersion in the unexposed area. By calculating the value, ie, the frequency characteristic (granularity noise) of the density unevenness in the unexposed area as the film characteristic, the previously recorded granularity noise characteristic of the film is selected according to the calculated dispersion value. This selection operation is performed only once when the film 2 is loaded.
[0045]
Here, the frequency characteristics are as described in the above-mentioned Non-Patent Document 1, pages 500 to 510, particularly the Wiener spectrum on page 506 and FIGS. 6.58 and 6.59 on page 508. And the type of film (film sensitivity, color film or black-and-white film, etc.).
[0046]
As described later with reference to FIG. 3, when the film noise estimating unit 9 determines that there is no corresponding granularity noise characteristic, the film noise estimating unit 9 determines the standard granularity noise characteristic. The noise characteristic is selected.
[0047]
Next, the film noise estimating unit 9 obtains the granularity noise amount using the average luminance value obtained from the luminance detecting unit 8 and the selected granularity noise characteristic, and calculates the obtained granularity noise amount by the noise reduction unit. Transfer to 10. The calculation of the granularity noise amount by the film noise estimation unit 9 is performed in synchronization with the processing of the noise reduction unit 10 based on the control of the control unit 13.
[0048]
The noise reduction unit 10 reads the video signal stored in the image buffer 7 based on the granularity noise amount acquired from the film noise estimation unit 9 and performs noise reduction processing in units of a predetermined size area. The subsequent video signal is transferred to the image processing unit 11.
[0049]
The image processing unit 11 performs known enhancement processing, compression processing, and the like on the video signal after noise reduction based on the control of the control unit 13.
[0050]
The output unit 12 records and stores the video signal processed by the image processing unit 11 on, for example, a disk-shaped recording medium.
[0051]
FIG. 2 is a block diagram illustrating an example of the configuration of the film noise estimating unit 9.
[0052]
The film noise estimating section 9 includes an unexposed area extracting section 21 serving as an extracting unit for extracting a video signal of an unexposed area in the video signal stored in the image buffer 7, and an unexposed area extracting section 21. A buffer 22 that stores the extracted video signal of the unexposed area, a variance calculating unit 23 that is a calculating unit that calculates the variance of the video signal of the unexposed area stored in the buffer 22 as a film characteristic, The relationship between the unexposed variance value and the granularity noise characteristic (see FIG. 14) is converted into a function, and the granularity noise ROM 25 serving as a granularity noise recording unit, which is stored in advance, and the variance calculating unit 23 calculates A selecting unit 24 serving as selecting means for selecting a function indicating the granularity noise characteristic stored in the granularity noise ROM 25 using the obtained variance value; When the function indicating the granularity noise characteristic is transferred from the controller, the function is transferred as it is to the noise amount calculation unit 27 described later, and when the function indicating the granularity noise characteristic is not transferred, a predetermined standard granularity noise is transferred. A standard value assigning unit 26 serving as an assigning unit for transferring a function indicating the characteristic to the noise amount calculating unit 27; a function indicating the granularity noise characteristic transferred from the standard value assigning unit 26; Based on the luminance value, the granularity noise amount is calculated in units of a region of a predetermined size (in this embodiment, a region of 2 × 2 pixels), and the calculated granularity noise amount for each region is sent to the noise reduction unit 10. And a noise amount calculating unit 27 for sequentially transferring the noise.
[0053]
The unexposed area extracting unit 21, the variance calculating unit 23, the selecting unit 24, and the standard value assigning unit 26 are bidirectionally connected to and controlled by the control unit 13.
[0054]
The operation of the film noise estimation unit 9 will be described with reference to FIG. FIG. 3 is a diagram for explaining an unexposed area.
[0055]
The film 2 includes an image area 2a where an image is captured, and an unexposed area 2b where an image is not captured. The imaging area 2c captured by the CCD 4 is, in this embodiment, , Including the image area 2a, so as to be wider than the image area 2a.
[0056]
Accordingly, the unexposed area extracting unit 21 extracts a video signal of the unexposed area 2b outside the image area 2a and inside the imaging area 2c. Which pixel position of the CCD 4 the unexposed area 2b to be extracted corresponds to can be set in advance according to the photographing magnification of the lens system 3.
[0057]
The unexposed area extracting unit 21 transfers the unexposed area extracted in this way to the buffer 22 for recording.
[0058]
The variance calculation unit 23 calculates a variance value from the video signal of the unexposed area stored in the buffer 22 under the control of the control unit 13, and transfers the calculated variance value to the selection unit 24.
[0059]
The selection unit 24 calculates the absolute value of the difference between the variance value acquired from the variance calculation unit 23 and all the variance values recorded in the granularity noise ROM 25 based on the control of the control unit 13. Then, the granularity noise characteristic corresponding to the variance value at which the absolute value of the obtained difference is minimized is selected, and the selected granularity noise characteristic is transferred to the standard value providing unit 26. However, when the minimum value of the absolute value of the difference calculated as described above exceeds a predetermined threshold, the selection unit 24 determines that the corresponding granularity noise characteristic is not recorded, and Does not forward anything to
[0060]
When a function indicating the granularity noise characteristic is transferred from the selection unit 24 based on the control of the control unit 13, the standard value providing unit 26 transfers this function to the noise amount calculation unit 27 as it is, and If the function indicating the noise characteristic is not transferred, the function is transferred to the noise amount calculation unit 27 to add a function indicating the predetermined standard granularity noise characteristic.
[0061]
The noise amount calculation unit 27 determines an area of a predetermined size based on the function indicating the granularity noise characteristic transferred from the standard value providing unit 26 and the brightness value transferred from the brightness detection unit 8 (this embodiment). In this embodiment, the granularity noise amount is calculated in units of (2 × 2 pixel regions), and the calculated granularity noise amount is sequentially transferred to the noise reduction unit 10 for each region.
[0062]
In the above description, the granularity noise characteristic is selected using the variance value of the unexposed area. However, the present invention is not limited to this. For example, by manually selecting the type of the film via the above-described film type selection operation unit of the external I / F unit 14, the granularity noise characteristic corresponding to the type of the film can be selected. It is also possible.
[0063]
The selection of the granularity noise characteristic by the selection unit 24 may be performed only once when the film 2 is loaded into the scanner system, but is not limited to this. For example, a change in the scene may be detected, and the granularity noise characteristic may be selected again each time, or an arbitrary time may be selected via the film type selection operation unit of the external I / F unit 14. May be selected.
[0064]
FIG. 4 is a block diagram illustrating an example of a configuration of the noise reduction unit 10.
[0065]
The noise reduction unit 10 obtains the granularity noise amount from the film noise estimating unit 9 and selects a filter size corresponding to the granularity noise amount. The size setting unit 34 is a filter setting unit. A local area extracting unit 31 that extracts a pixel block corresponding to the filter size set by the size setting unit 34 from the image buffer 7 and a filtering unit in which a preset coefficient is recorded corresponding to the filter size. A coefficient ROM 35 and a coefficient corresponding to the filter size acquired from the size setting unit 34 are read from the coefficient ROM 35, and a known smoothing is performed on the video signal of the pixel block extracted by the local area extracting unit 31. 32 as filtering means for performing a filtering process for It is configured to have a buffer 33 for storing the video signal processed by the filtering unit 32, a.
[0066]
The local area extracting unit 31, the filtering unit 32, and the size setting unit 34 are bidirectionally connected to and controlled by the control unit 13.
[0067]
Under the control of the control unit 13, the noise reduction unit 10 repeatedly performs the noise reduction process for each pixel block, so that all the signals of the image buffer 7 are finally processed and stored in the buffer 33. Become. The image processing unit 11 reads out the video signal from the buffer 33 and processes it after the filtering process for all the signals is completed.
[0068]
The filter size set by the size setting unit 34 corresponds to the pigment cloud size, for example, as shown in FIG. 5, that is, the filter size corresponds to the granularity noise caused by the pigment cloud. FIG. 5 is a diagram illustrating an example of the relationship between the granularity noise amount and the filter size.
[0069]
The size setting unit 34 sets the smoothing filter size of 3 × 3 pixels when the pigment cloud size is 3 to 10 (μm), and sets the smoothing filter size when the pigment cloud size is 11 to 20 (μm). A 5 × 5 pixel smoothing filter size is selected, and a 7 × 7 pixel smoothing filter size is selected when the pigment cloud size is 21 to 30 (μm).
[0070]
Although the filtering process in the present embodiment is an example of a smoothing process having no directivity, the filtering process need not be limited to this. For example, if the granularity noise has directionality depending on the type of film, it is also possible to detect the directionality by known edge extraction or the like and perform an asymmetric filtering process in consideration of the detected directionality.
[0071]
In the above description, the processing is performed by hardware, but the present invention is not limited to this, and the processing may be performed by software. For example, the signal from the CCD 4 is used as unprocessed raw data, film type information from the film noise estimating unit 9 is added as header information of the raw data, and the raw data with the header information is added. It is conceivable to perform processing by a computer using separate software.
[0072]
FIG. 6 is a flowchart illustrating software processing for noise reduction.
[0073]
When this process is started, a video signal obtained as Raw data and header information such as film type information are read into a computer (step S1).
[0074]
The film type information is extracted from the read information and transferred to step S6 described below (step S2).
[0075]
The video signals read in step S1 are sequentially scanned (step S3).
[0076]
A region of a predetermined size around the target pixel, for example, a region of 2 × 2 pixels is extracted (step S4), an average luminance value of the extracted region is calculated, and transferred to step S6 described later (step S5).
[0077]
Based on the film type information extracted in step S2, the separately recorded granularity noise characteristic is read, and the granularity noise amount is obtained using the luminance value calculated in step S5 (step S6).
[0078]
The filtering size is obtained from the granularity noise amount obtained in step S6, and is transferred to steps S8, S9, and S10 described later (step S7).
[0079]
An area corresponding to the filtering size obtained in step S7 is extracted around the pixel of interest, and is transferred to step S10 described later (step S8).
[0080]
In addition, a coefficient corresponding to the filtering size obtained in step S7 is read and transferred to step S10 described below (step S9).
[0081]
The region extracted in step S8 is subjected to smoothing filtering processing using the coefficient transferred from step S9 (step S10), and the smoothed signals are sequentially output (step S11).
[0082]
Then, it is determined whether or not the scanning has been completed for one screen (step S12). If the scanning has not been completed, the process returns to step S3 to repeat the above-described processing. Ends.
[0083]
According to the first embodiment, the granularity noise amount is estimated in accordance with the type of the film and the luminance value level, and the smoothing process in which the filtering size is changed based on the estimated granularity noise amount is performed. Therefore, it is possible to obtain a high-quality image in which only the granularity noise component is removed with high accuracy. At this time, since the estimation of the granularity noise amount is performed in a predetermined area as a unit, it is possible to perform optimal noise reduction from a bright portion to a dark portion.
[0084]
In addition, the operability is improved because the granularity noise characteristics of each film are recorded in advance and the granularity noise characteristics are automatically selected based on the information of the unexposed area of the film when used. Can be. At this time, as the information on the unexposed area, the variance of the unexposed area having a small amount of calculation and a high correlation with the noise characteristic is used, it is possible to realize a high-performance scanner system at low cost. it can.
[0085]
On the other hand, when the granularity noise characteristic of each film is recorded in advance and the granularity noise characteristic is selected by manually selecting the film type at the time of use, the optimal noise characteristic is selected to achieve high quality. Images can be obtained, and the cost of the entire system can be reduced.
[0086]
Further, when the granularity noise characteristic corresponding to the information of the designated film or the unexposed area is not recorded, the standard granularity noise characteristic is added and used, so that the granularity noise characteristic is The noise reduction process can be performed even on a type of film that has not been recorded in advance, and a highly versatile scanner system can be obtained.
[0087]
Since the noise reduction process is performed by the filtering process, the operation can be performed with a small-sized memory, and the cost can be reduced.
[0088]
7 to 13 show a second embodiment of the present invention, and FIG. 7 is a block diagram showing a configuration of a scanner system. In the second embodiment, the same components as those in the above-described first embodiment are denoted by the same reference numerals, and description thereof will be omitted. Only different points will be mainly described.
[0089]
The scanner system according to the second embodiment is obtained by adding an element noise estimating unit 41 as element noise estimating means to the configuration shown in FIG. 1 of the first embodiment.
[0090]
The element noise estimating section 41 estimates the element noise amount of the imaging system including the CCD 4 based on the luminance value detected by the luminance detecting section 8 and outputs the estimated element noise amount to the noise reduction section 10. ing.
[0091]
This element noise estimating unit 41 is also bidirectionally connected to the control unit 13 so as to be controlled.
[0092]
Next, the operation of the scanner system as shown in FIG. 7 is basically the same as that of the above-described first embodiment, and therefore only different parts will be described.
[0093]
The element noise estimating unit 41 is configured to control the CCD 4 or the amplifier 5 based on the element noise characteristics recorded in advance and the average luminance value of the area of, for example, 2 × 2 pixels detected by the luminance detecting unit 8. The amount of element noise caused by the imaging system is obtained for each area and transferred to the noise reduction unit 10.
[0094]
The noise reduction unit 10 obtains a total noise amount by adding the granularity noise amount from the film noise estimation unit 9 and the element noise amount from the element noise estimation unit 41, and the total noise amount is stored in the image buffer 7. A noise reduction process is performed on a video signal in a region of a predetermined size as a unit.
[0095]
The processing of the noise reduction unit 10 is synchronized with the calculation of the element noise amount by the element noise estimation unit 41 and the calculation of the granularity noise amount by the film noise estimation unit 9 based on the control of the control unit 13. Is being done.
[0096]
The transfer of the video signal from the noise reduction unit 10 to the image processing unit 11 for processing is the same as in the first embodiment described above.
[0097]
FIG. 8 is a block diagram illustrating an example of the configuration of the element noise estimation unit 41.
[0098]
The element noise estimating unit 41 includes an element noise ROM 52 serving as element noise recording means in which element noise characteristics are recorded in advance as described later with reference to FIG. 9 and a video signal read from the control unit 13. A gain correction unit 51 serving as gain correction means for correcting the element noise characteristic by multiplying the gain amount of the amplifier 5 by the element noise characteristic read from the element noise ROM 52; And a noise amount calculation unit 53 that obtains an element noise amount corresponding to the luminance value from the luminance detection unit 8 based on the noise characteristics and transfers the element noise amount to the noise reduction unit 10.
[0099]
The gain correction unit 51 is bidirectionally connected to the control unit 13 and controlled.
[0100]
FIG. 9 is a diagram showing how the element noise changes with respect to the luminance value.
[0101]
As shown in the example of FIG. 9, the element noise characteristics recorded in the element noise ROM 52 are such that the noise amount N increases to the power L with respect to the luminance value L, for example. Such element noise characteristics can be easily obtained by measuring the characteristics of the imaging element system in advance.
[0102]
FIG. 10 is a block diagram illustrating a first configuration example of the noise reduction unit 10.
[0103]
The noise reduction unit 10 shown in FIG. 10 is obtained by adding an addition unit 61 as addition means to the configuration shown in FIG. 4 of the first embodiment.
[0104]
The adding unit 61 adds the granularity noise amount acquired from the film noise estimating unit 9 and the element noise amount acquired from the element noise estimating unit 41 to obtain a total noise amount.
[0105]
The adding unit 61 is bidirectionally connected to and controlled by the control unit 13 together with the local region extracting unit 31, the filtering unit 32, and the size setting unit 34.
[0106]
Also, unlike the first embodiment described above, the size setting unit 34 selects a filter size for the total noise amount transferred from the adding unit 61, and selects the filter size for the local region extracting unit 31 and the filtering unit 32. Output to
[0107]
According to the first configuration example, it is possible to construct a scanner system that integrally reduces the granularity noise of the film and the device noise of the image sensor, and obtains a high-quality image at low cost. It becomes.
[0108]
In the first example of the noise reduction processing, the filtering unit 32 performs the filtering based on the frequency characteristics. However, the present invention is not limited to this. It is also possible to apply means for correcting the video signal.
[0109]
FIG. 11 is a block diagram illustrating a second configuration example of the noise reduction unit 10.
[0110]
The noise reduction unit 10 includes an average value calculation unit 73 that sequentially extracts local regions of a predetermined size from the image buffer 7 and calculates an average value of video signals for each region. An adding unit 71 serving as an adding unit for adding the obtained granularity noise amount and the element noise amount obtained from the element noise estimating unit 41 to calculate a total noise amount; and an integrated unit transferred from the adding unit 71. An allowable range setting unit 72 as an allowable range setting unit that sets an upper limit value Tup and a lower limit value Tlow for the video signal based on the appropriate noise amount and the average value transferred from the average value calculating unit 73, Based on the upper limit value Tup and the lower limit value Tlow from the allowable range setting section 72, the image of the area corresponding to the local area used in the average value calculating section 73 from the image buffer 7 described above. It is configured to include a correction means serving correcting unit 74 to be outputted to the image processing unit 11 sequentially extracted and correction processing result items.
[0111]
The adding unit 71, the allowable range setting unit 72, the average value calculating unit 73, and the correcting unit 74 are bidirectionally connected to and controlled by the control unit 13, and the allowable range setting unit 72 among them is controlled. The average value calculation unit 73 and the correction unit 74 are controlled so as to perform processing in synchronization with each local area.
[0112]
Based on the total noise amount N from the adding unit 71 and the average value AV from the average value calculating unit 73, the allowable range setting unit 72 sets an upper limit value Tup and a lower limit value Tlow for the video signal. Is set according to the following equation 1.
(Equation 1)
Tup = AV + N / 2
Tlow = AV-N / 2
[0113]
Further, the correcting unit 74 selects three types of processing as shown in the following Expression 2 from the upper limit value Tup and the lower limit value Tlow for each pixel S of the extracted local region, and determines the element S. to correct.
(Equation 2)
S = S−N / 2 (when Tup <S)
S = AV (when Tlow ≦ S ≦ Tup)
S = S + N / 2 (when S <Tlow)
[0114]
If the element noise estimating unit 41 and the adding unit 71 are omitted, the second configuration example can be applied to the above-described first embodiment.
[0115]
The noise reduction process can be performed by, for example, a smoothing process using a threshold, in addition to the first configuration example shown in FIG. 10 and the second configuration example shown in FIG.
[0116]
FIG. 12 is a block diagram illustrating a third configuration example of the noise reduction unit 10.
[0117]
The noise reduction unit 10 includes a horizontal line extraction unit 81 that sequentially extracts video signals from the image buffer 7 in units of horizontal lines, a granularity noise amount acquired from the film noise estimation unit 9 and the element noise estimation unit 41. The addition unit 86 is an addition unit that calculates the total noise amount by adding the element noise amount obtained from the above. The addition unit 86 applies the horizontal line unit video signal extracted by the horizontal line extraction unit 81 to the addition unit 86. The acquired total noise amount is set as a threshold value and transferred to the first smoothing unit 82, and the overall signal from the adding unit 86 is applied to the vertical line unit video signal extracted by the vertical line extracting unit 84 described later. A threshold setting unit 87 that sets the noise amount as a threshold and transfers it to a second smoothing unit 85, which will be described later, and a video signal from the horizontal line extraction unit 81 A first smoothing unit 82 as a smoothing means for sequentially performing scanning in elementary units and performing a known hysteresis smoothing using the threshold value from the threshold value setting unit 87 as a noise amount, and sequentially accumulating the output of the first smoothing unit 82 and finally storing A buffer 83 for storing video signals for one screen, a vertical line extracting unit 84 for sequentially extracting video signals in units of vertical lines from the video signals for one screen stored in the buffer 83, and a vertical line extracting unit 84 And a second smoothing unit 85 as smoothing means for sequentially scanning the video signal from the pixel unit in pixel units, performing a known hysteresis smoothing using the threshold value from the threshold value setting unit 87 as a noise amount, and sequentially outputting the noise value to the image processing unit 11. It is configured.
[0118]
The horizontal line extracting section 81, the vertical line extracting section 84, the adding section 86, and the threshold setting section 87 are bidirectionally connected to and controlled by the control section 13, and the adding section 86 among them is provided. The operation of the threshold setting unit 87 and the hysteresis smoothing operation of the first smoothing unit 82 are controlled to be synchronized with each other, and the operation of the addition unit 86 and the threshold setting unit 87 and the hysteresis smoothing of the second smoothing unit 85 are also controlled. Controlled to synchronize.
[0119]
In such a configuration, horizontal smoothing is performed by the first smoothing unit 82, processing of all signals stored in the image buffer 7 ends, and the processed video signal for one screen is stored in the buffer 83. After that, vertical smoothing by the second smoothing section 85 is performed.
[0120]
If the element noise estimating unit 41 and the adding unit 86 are omitted, the third configuration example can be applied to the above-described first embodiment.
[0121]
Further, in the above description, the processing is performed by hardware, but the processing is not limited to this, and the processing may be performed by software. For example, the signal from the CCD 4 is added as unprocessed Raw data, and the film type information from the film noise estimating unit 9 and the gain amount from the control unit 13 are added as header information, and this header information is added. The RAW data may be processed by a computer using separate software.
[0122]
FIG. 13 is a flowchart illustrating noise reduction software processing. The software processing shown in FIG. 13 performs the same processing as the noise reduction unit 10 having the configuration shown in FIG. 11 as software.
[0123]
When this process is started, a video signal obtained as Raw data and header information such as film type information and gain amount are read into a computer (step S21).
[0124]
The gain information is extracted from the read information and transferred to step S28 described later (step S22), and the film type information is extracted and transferred to step S27 described later (step S23).
[0125]
The video signals read in step S21 are sequentially scanned (step S24).
[0126]
An area of a predetermined size, for example, an area of 2 × 2 pixels is extracted around the pixel of interest (step S25), an average luminance value of the extracted area is calculated, and the average luminance value is transferred to steps S27, S28, and S30 described below ( Step S26).
[0127]
Based on the film type information extracted in step S23, the separately recorded granularity noise characteristic is read, and the granularity noise amount is obtained using the luminance value calculated in step S26 (step S27).
[0128]
In addition, the device noise characteristic separately recorded is read, the correction is performed based on the gain amount from step S22, and the device noise amount is obtained based on the luminance value from step S26 (step S28).
[0129]
The granularity noise amount from step S27 is added to the element noise amount from step S28, and the addition result is transferred to step S30 as a total noise amount (step S29).
[0130]
Using the average luminance value AV from step S26 and the total noise amount N from step S29, the allowable ranges Tup and Tlow as shown in the above equation 1 are calculated (step S30).
[0131]
On the other hand, a local area of a predetermined size is extracted from the video signal scanned in step S24 and transferred to step S32 (step S31).
[0132]
Using the permissible ranges Tup, Tlow and the average luminance value AV from step S30, the correction processing as shown in the above equation 2 is performed on each pixel in the local area (step S32).
[0133]
After sequentially outputting the corrected signals (step S33), it is determined whether or not the scanning has been completed for one screen (step S34). Are repeated, and if the scanning is completed, the process is terminated.
[0134]
Here, an example in which the processing of the noise reduction unit 10 shown in FIG. 11 is performed by software has been described. However, the processing of the noise reduction unit 10 as shown in FIGS. 10 and 12 may be performed by software. Of course it is possible.
[0135]
According to the second embodiment, substantially the same effects as those of the first embodiment described above can be obtained, and not only the granularity noise caused by the film but also the device noise caused by the image sensor can be effectively reduced. And a high-quality image can be obtained. At this time, since the estimation of the granularity noise amount and the element noise amount is performed in a predetermined area as a unit, it is possible to perform optimal noise reduction from a bright portion to a dark portion.
[0136]
Further, since both the granularity noise and the element noise are reduced by the same noise reduction unit, the configuration can be simplified and the cost can be reduced as compared with the case where each is processed separately.
[0137]
Further, since the device noise amount is obtained after correcting the device noise characteristics of the image pickup device based on the gain amount for the video signal, more appropriate noise reduction processing according to the gain amount can be performed.
[0138]
Further, since a filter size corresponding to the granularity noise amount and the element noise amount is selected and the noise reduction processing is performed using the filter size, only the components of both the granularity noise and the element noise are removed. And high-quality images can be obtained. In this case, since the noise reduction processing is filtering processing, work can be performed with a small-sized memory, and cost reduction can be achieved. Further, since the amount of calculation is small in the correction processing by filtering, high-speed processing becomes possible.
[0139]
On the other hand, a high-quality image can also be obtained by setting the upper limit and the lower limit and removing the granularity noise component. The correction processing using the upper limit value and the lower limit value can be performed at high speed because the amount of calculation is small.
[0140]
Then, the granularity noise amount (or the granularity noise amount and the element noise amount) is set as a threshold, and when the signal below this threshold value is removed as noise, the granularity noise (or the granularity noise and the element noise) is also determined. Only both components are removed, and a high-quality image can be obtained. Since this smoothing process can perform flexible control, it is possible to configure a scanner system applicable to various noises.
[0141]
It should be noted that the present invention is not limited to the above-described embodiment, and it is needless to say that various modifications and applications can be made without departing from the gist of the invention.
[0142]
【The invention's effect】
As described above, according to the scanner system and the image processing program of the present invention, it is possible to obtain a high-quality image in which at least one of the granularity noise and the element noise is favorably reduced without depending on the luminance change.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a scanner system according to a first embodiment of the present invention.
FIG. 2 is a block diagram illustrating an example of a configuration of a film noise estimating unit according to the first embodiment.
FIG. 3 is a diagram for explaining an unexposed area in the first embodiment.
FIG. 4 is a block diagram illustrating an example of a configuration of a noise reduction unit according to the first embodiment.
FIG. 5 is a diagram showing an example of a relationship between a granularity noise amount and a filter size in the first embodiment.
FIG. 6 is a flowchart illustrating noise reduction software processing according to the first embodiment.
FIG. 7 is a block diagram illustrating a configuration of a scanner system according to a second embodiment of the present invention.
FIG. 8 is a block diagram illustrating an example of a configuration of an element noise estimating unit according to the second embodiment.
FIG. 9 is a diagram showing a state of change of element noise with respect to a luminance value in the second embodiment.
FIG. 10 is a block diagram showing a first configuration example of a noise reduction unit in the second embodiment.
FIG. 11 is a block diagram showing a second configuration example of the noise reduction unit in the second embodiment.
FIG. 12 is a block diagram showing a third configuration example of the noise reduction unit in the second embodiment.
FIG. 13 is a flowchart illustrating software processing for noise reduction according to the second embodiment.
FIG. 14 is a diagram showing a typical example of how RMS granularity changes with film density.
[Explanation of symbols]
1. Light source
2. Film
2a: Image area
2b: unexposed area
2c: imaging area
3 ... Lens system
4: CCD (image sensor)
5 ... Amplifier
6 A / D converter
7 ... Image buffer
8. Luminance detection unit (luminance detection means)
9 ... Film noise estimation unit (film noise estimation means)
10. Noise reduction unit (noise reduction means)
11 Image processing unit
12 Output unit
13: control unit (operation means)
14. External I / F unit (operation means)
21 ... Unexposed area extraction unit (extraction means)
22, 33, 83 ... buffer
23 variance calculation unit (calculation means)
24 ... Selection unit (selection means)
25: ROM for granularity noise (granularity noise recording means)
26 ... Standard value assigning unit (assigning means)
27, 53: Noise amount calculation unit
31 local area extracting unit
32 ... Filtering unit (filtering means)
34: Size setting section (filter setting means)
35 ... Coefficient ROM (filtering means)
41: Element noise estimation unit (element noise estimation means)
51: gain correction unit (gain correction means)
52: ROM for element noise (element noise recording means)
61, 71, 86... Adder (adder)
72 ... Permissible range setting unit (permissible range setting means)
73 ... Average value calculation unit
74. Correction unit (correction means)
81: Horizontal line extraction unit
82 first smoothing unit (smoothing means)
84 vertical line extraction unit
85... Second smoothing part (smoothing means)
87: threshold setting unit (threshold setting means)

Claims (15)

A scanner system that captures a video signal of an image on the film by capturing an image of the developed film with an imaging device,
Brightness detection means for detecting brightness information for each region of a predetermined size from the video signal,
Film noise estimation means for estimating the granularity noise amount of the film for each region based on the luminance information detected by the luminance detection means,
Noise reduction means for reducing the noise of the video signal based on the granularity noise amount estimated by the film noise estimation means,
A scanner system comprising: A scanner system that captures a video signal of an image on the film by capturing an image of the developed film with an imaging device,
Brightness detection means for detecting brightness information for each region of a predetermined size from the video signal,
Film noise estimation means for estimating the granularity noise amount of the film for each region based on the luminance information detected by the luminance detection means,
Element noise estimating means for estimating the element noise amount of the image sensor for each of the regions based on the luminance information detected by the luminance detecting means,
Noise reduction means for reducing the noise of the video signal based on the granularity noise amount estimated by the film noise estimation means and the element noise amount estimated by the element noise estimation means,
A scanner system comprising: The film noise estimating means includes:
Operating means for selecting the type of the film,
A granularity noise recording means for recording granularity noise characteristics of a plurality of types of films,
Selecting means for selecting the granularity noise characteristic recorded in the granularity noise recording means based on the type of film selected by the operating means,
Is configured with
3. The scanner system according to claim 1, wherein a granularity noise amount of the film is estimated from the luminance information based on the granularity noise characteristic selected by the selection unit. The film noise estimating means includes:
Extracting means for extracting a video signal of an unexposed area from the video signal,
Calculation means for calculating film characteristics based on the video signal of the unexposed area extracted by the extraction means,
A granularity noise recording means for recording granularity noise characteristics of a plurality of types of films,
Selecting means for selecting the granularity noise characteristic recorded in the granularity noise recording means based on the film characteristic calculated by the calculating means,
Is configured with
3. The scanner system according to claim 1, wherein a granularity noise amount of the film is estimated from the luminance information based on the granularity noise characteristic selected by the selection unit. The film noise estimating means includes:
The apparatus further includes an applying unit that applies a predetermined granularity noise characteristic,
When it is not possible to select the granularity noise characteristic recorded in the granularity noise recording means by the selection means, the film information is obtained from the luminance information based on the predetermined granularity noise characteristic provided by the provision means. 5. The scanner system according to claim 3, wherein the granularity noise amount is estimated. 6. 5. The scanner system according to claim 4, wherein the calculating means calculates a variance of the video signal in the unexposed area, and calculates a film characteristic based on the variance. The element noise estimating means includes:
Element noise recording means for recording element noise characteristics of the image pickup element,
Gain correction means for reading element noise characteristics from the element noise recording means and correcting based on a gain for a video signal,
Is configured with
3. The scanner system according to claim 2, wherein an element noise amount of the image pickup element is estimated for each of the regions based on the element noise characteristics corrected by the gain correction unit and the luminance information. The noise reduction means includes:
Filter setting means for setting a filter characteristic based on the granularity noise amount,
Filtering means for reducing noise by filtering the video signal using the filter characteristics set by the filter setting means,
The scanner system according to claim 1, wherein the scanner system is configured to have: The noise reduction means includes:
Adding means for adding the granularity noise amount and the element noise amount,
Filter setting means for setting a filter characteristic based on the noise amount added by the adding means,
Filtering means for reducing noise by filtering the video signal using the filter characteristics set by the filter setting means,
The scanner system according to claim 2, wherein the scanner system is configured to have: The noise reduction means includes:
An allowable range setting means for setting an upper limit and a lower limit based on the granularity noise amount and the luminance information,
Correction means for reducing noise by correcting the video signal based on the upper limit value and the lower limit value set by the allowable range setting means,
The scanner system according to claim 1, wherein the scanner system is configured to have: The noise reduction means includes:
Adding means for adding the granularity noise amount and the element noise amount,
Allowable range setting means for setting an upper limit and a lower limit based on the noise amount and the luminance information added by the adding means,
Correction means for reducing noise by correcting the video signal based on the upper limit value and the lower limit value set by the allowable range setting means,
The scanner system according to claim 2, wherein the scanner system is configured to have: The noise reduction means includes:
Threshold setting means for setting a noise amplitude value as a threshold based on the granularity noise amount,
Smoothing means for reducing the amplitude component of the video signal which is equal to or less than the threshold value set by the threshold value setting means,
The scanner system according to claim 1, wherein the scanner system is configured to have: The noise reduction means includes:
Adding means for adding the granularity noise amount and the element noise amount,
Threshold setting means for setting a noise amplitude value as a threshold based on the noise amount added by the adding means,
Smoothing means for reducing the amplitude component of the video signal which is equal to or less than the threshold value set by the threshold value setting means,
The scanner system according to claim 2, wherein the scanner system is configured to have: Computer
Brightness detection means for detecting brightness information for each region of a predetermined size from a video signal relating to an image on the film captured by capturing an image of the developed film by the imaging device,
Film noise estimating means for estimating the granularity noise amount of the film for each area based on the luminance information detected by the luminance detecting means,
A noise reduction unit that reduces noise of the video signal based on the granularity noise amount estimated by the film noise estimation unit;
An image processing program characterized by functioning as a computer. Computer
Brightness detection means for detecting brightness information for each region of a predetermined size from a video signal relating to an image on the film captured by capturing an image of the developed film by the imaging device,
Film noise estimating means for estimating the granularity noise amount of the film for each area based on the luminance information detected by the luminance detecting means,
Element noise estimating means for estimating the element noise amount of the image sensor for each of the regions based on the luminance information detected by the luminance detecting means;
A noise reduction unit that reduces noise of the video signal based on the granularity noise amount estimated by the film noise estimation unit and the element noise amount estimated by the element noise estimation unit;
An image processing program characterized by functioning as a computer.

Images