JP2001333326A - Digital camera having image processing function - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a digital camera having an image processing function that eliminates the need for a sensor detecting any physical quantity for correction factors such as camera-shake so as to obtain an excellent image. SOLUTION: The digital camera having the image processing function that corrects an image and provided with a storage device 101 that stores a picked-up image and a display device 104 that displays the image, is provided with an image processing means 102 that corrects the picked-up image, and an operation means 103 that gives correction information to the processing means. The operation means is characteristically configured that is operated externally and the operation means properly provides correction information to the picked-up image to correct the image on the basis of the image that is displayed on the display device and picked up.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a digital camera having an image processing function to which a function of restoring an image deteriorated due to camera shake or defocus is added.

[0002]

2. Description of the Related Art Hitherto, various mechanisms for correcting image deterioration due to camera shake in a digital camera have been proposed. For example, as shown in FIG. 7, an optical system such as a lens and a light receiving element such as a CCD are shown in FIG. The imaging unit 500 configured by
Although it differs depending on the method of correcting camera shake, it is configured to include a mechanical mechanism for correcting vibration of the optical system or an arithmetic circuit for converting pixel information of the CCD.

According to such a conventional technique, when a camera is vibrated due to hand shake during photographing, the amount of the vibration is detected by a vibration detection sensor 501. The detected vibration is
The vibration correction arithmetic circuit 502 calculates a motion vector of the camera itself and obtains a correction value. Based on the correction value, the optical system is mechanically corrected, or the value of each pixel of the light receiving element is corrected by image processing calculation, thereby preventing or correcting image deterioration such as camera shake. as a result,
An image for which camera shake has been corrected is recorded on a recording medium 101 composed of, for example, a flash memory or a hard disk.

[0004] Each of these prior arts includes the following two characteristic elements. (1) A sensor that detects some physical quantity for correction factors such as camera shake is used. (2) When a correction such as a camera shake is necessary, the correction is performed by some correction means, and the corrected image is recorded on a recording medium.

[0005]

According to the above prior art, from the above points, there are the following problems in the method of correcting image deterioration due to camera shake or the like in a conventional digital camera. ○ A sensor that detects some physical quantity is required for correction factors such as camera shake. If an accurate correction amount is not obtained for correction of camera shake or the like, an image of an incorrect correction result is recorded on a recording medium.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has no need for a sensor for detecting any physical quantity for correction factors such as camera shake, and has a digital image processing function capable of obtaining a good image. The purpose is to provide a camera. Another object of the present invention is to prevent an erroneous correction result from being recorded on a recording medium.

[0007]

SUMMARY OF THE INVENTION The present invention provides a digital camera having a storage device for storing a captured image and a display device for displaying the image, and having an image processing function for correcting the image. Image processing means for correcting the image, and operating means for providing correction information to the image processing means, wherein the operating means is configured to be operable from the outside, and based on a captured image displayed on the display device. ,
The image is corrected by appropriately adding correction information by the operation means.

It is desirable to use a degraded image restoration method by a blind deconvolution method as the image processing function. According to this technology, an operation unit that provides correction information to an image processing unit that corrects a captured image is configured to be operable from the outside, and based on the captured image displayed on the display device, the operation unit Since the image is corrected by appropriately adding correction information, there is no need to provide a mechanism for automatically detecting a blur amount for the camera shake image, and the camera shake image or out-of-focus image can be corrected with a simple configuration. .

Also, since a mechanism for automatically detecting the amount of blur is not used, correction can be normally made along the subject image picked up by the operator, and an image of an incorrect correction result is recorded on a recording medium. Can be prevented.

In addition, since the image processing apparatus has an image processing function based on a degraded image restoration method based on a blind deconvolution method, the processed image recording medium can be detached without using an image restoration process using a computer such as a personal computer. By using a recording medium or the like, it is possible to directly set the image recording medium on an external output device or the like and output a good image in which camera shake and the like are restored.

Further, the storage device may further comprise an uncorrected image storage means for storing the image before correction and an after-correction image storage means for storing the corrected image. It is. According to such a technical means, a recording medium for recording both a raw image before correction and a corrected image after image processing is provided for an image to be photographed. When the amount is measured and corrected, even if the amount of incomplete blur is measured, no further correction is possible, and correction that deviates extremely from the raw image can be prevented. Recording of the image of the correction result can be prevented.

The storage device includes image correction information storage means for storing a history of image correction information,
It is also an effective means of the present invention to make the previous history usable by the operation means.

According to such technical means, the prior history of the image correction information which is a cause of deterioration is appropriately recorded and read out, and when restoring another deteriorated image, the previous history data which is recorded is suitable. An image can be selected and used, and this has the effect of making it possible to reduce the degradation image restoration processing time.

[0014]

Embodiments of the present invention will be illustratively described in detail below with reference to the drawings. However, unless otherwise specified, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention, but are merely illustrative examples. It's just

FIG. 1 is a block diagram of a digital camera having an image processing function according to a first embodiment of the present invention, and FIG.
Content of the image processing function according to the embodiment of the present invention,
FIG. 3 is a diagram showing an image restoration algorithm by the Fourier iteration method, FIG. 3 is a model diagram of a non-negative condition of the image restoration algorithm by the Fourier iteration method, FIG. 4 is a configuration diagram of an image processing operation circuit according to the embodiment of the present invention; FIG. 5 is a schematic diagram of the arithmetic processing of the image processing arithmetic circuit according to the embodiment of the present invention, and FIG. 6 is a configuration diagram of a digital camera having an image processing function according to the second embodiment of the present invention.

In FIG. 1, an image pickup unit 100 includes an optical system such as a lens ゛ and a light receiving element such as a CCD. Recording medium 101 for recording image information received by imaging unit 100
Is for recording a raw image before correction, and is composed of a flash memory, but may be a hard disk or the like. An image output unit 104 capable of observing an image recorded on the recording medium 101 is constituted by a liquid crystal display.

The image processing operation circuit 102 is a method of restoring a deteriorated image by a blind deconvolution method (for example,
It has a function of an image restoration method by a Fourier iteration method) and includes, for example, an operation image storage memory, a Fourier transform circuit, an arithmetic operation circuit, and the like. The degraded image restoration method by the blind deconvolution method is described in, for example, the literature (GRAyers and JCDainty, Opt. Lett. 13,
547-549 [1988]), etc. This is an image restoration method using the Fourier iterative method.
This is a method of separating a restored image that has not deteriorated and a deterioration factor.

The input operation unit 103 includes switches, a touch panel, and the like. When the observer determines that the image observed by the image output unit 104 has deteriorated due to camera shake, etc., the image restoration processing is performed. Start and stop operations are configured to be possible. Also, the processed image recording medium 105
Is a recording medium for recording a processed image subjected to the restoration processing by the image processing operation circuit 102.

Here, as a method of restoring a deteriorated image by the blind deconvolution method, an image restoration method by the Fourier iteration method will be described. A deteriorated image due to camera shake or the like is mathematically represented by the following.

(Equation 1) c (x, y): blurred image to be restored, f (ξ, η): unblurred original image, g (x−ξ, y−η): point image response function as a blur factor

In the equation (1), the blurred image c is a convolution of the original image f and the point image response function g.
It is represented by When the Fourier transforms of c, f, and g are represented by C, F, and G, respectively, in the spatial frequency domain (Fourier transform plane), C (μ, ν) = F (μ, ν) · G (μ, ν) ) (2)

If the blurred image c and the point image response function g are known for the blurred image represented by the formula (1) or (2), the original image f can be easily restored by the following formula. Can be. F (μ, ν) = C (μ, ν) / G (μ, ν) (3)

However, there are few examples in which the point image response function g is accurately known in the image restoration processing, and in many cases, only the blurred image c to be restored is known.
The restoration processing by Expression (3) is called deconvolution, and is a technique widely used not only for image processing but also for signal processing in general. On the other hand, from only the blurred image c, the expression (2)
The method of estimating and separating F and G in is referred to as a blind deconvolution method.

Correction of a deteriorated image such as camera shake in the prior art is nothing more than detecting the value of G (μ, υ) using various sensors and performing deconvolution according to equation (3). However, the degraded image restoration method by the blind deconvolution method in the present embodiment, as described later, estimates the f and g while performing an iterative calculation while gazing at the blurred image c on the liquid crystal display. This is a technique that converges to a relationship satisfying (2).

FIG. 2 shows an algorithm of the image restoration technique by the Fourier iteration method. In FIG. 2, the estimated original image f, the estimated point image response function g, the initial original image f0, the Fourier transform C of the blurred image, the Fourier transform F of the estimated original image, the Fourier transform G of the estimated point image response function, and the estimated original image f and the Fourier transforms F and G of the estimated point image response function g, and the inverse Fourier transforms that return them (Step 3) and (Step 7), and deconvolution (Step 6) which is the restoration processing shown in Expression (3) ), And by performing iterative calculation while giving non-negative conditions (step 4) and (step 8), the relationship between F and G satisfying the expression (2) is converged.

The non-negative condition refers to the operation of converting an implicit negative value into each pixel of the image information f and g obtained in the iterative calculation into a positive value. FIG. 3 models this non-negative condition. The horizontal axis represents each pixel position (for example, the position of each light receiving unit with respect to a CCD camera in which the light receiving units are arranged in a two-dimensional matrix. In FIG. 3, for simplicity of description, the light receiving units are arranged in a line). The vertical axis represents the light intensity.

FIG. 3A shows image information in which a negative value is generated by iterative calculation, and FIG. 3B shows how a negative value is converted to a positive value as a non-negative condition. As a condition for converting to a positive value, a total energy conservation law needs to be satisfied before and after a non-negative condition. Here, for example, an operation is shown in which the total sum of the energy amounts that have produced negative values is divided by the number of pixels of the entire image, and the average is added to each pixel.

Next, the operation and effects of the first embodiment will be described with reference to FIG. Optical system such as lens and CC
Image information received by the imaging unit 100 including light receiving elements such as D is recorded on the recording medium 101. Recording medium 1
The image recorded in 01 is observed on the liquid crystal panel of the image output unit 104. Here, the image recorded on the recording medium 101 is recorded as a deteriorated image even when a deterioration factor such as camera shake exists.

When the observer determines that the image observed by the image output unit 104 has deteriorated due to camera shake or the like, an image restoration operation is performed using switches, a touch panel, or the like in the input operation unit 103. Will be The restoration processing of the deteriorated image by the blind deconvolution method is performed by the iterative calculation as shown in FIG. 2 as described above. In this iterative calculation, C = FG in equation (2), (C is the Fourier transform of the degraded image, F is the Fourier transform of the original image, G
Represents the Fourier transform of the point spread response function that causes deterioration.)
, C is known in the obtained deteriorated image. To start the iterative calculation, the initial value of F or G is estimated. Normally, F and G are completely unknown, so random numbers are used as initial values. Here, for example, the initial value is F of a random number. G for this initial value is obtained from the relationship of equation (2). For this G, as shown in FIG.
A non-negative constraint is applied to make it closer to the correct answer.
Using the G approaching the correct answer, F is obtained from the relationship of equation (2). Similar to G, non-negative constraint conditions are applied to this F to make it closer to the correct answer. By repeating this operation by iterative calculation, both F and G approach the correct answer.

In the method of restoring a deteriorated image by the blind deconvolution method, the convergence process is performed by iterative calculation. Therefore, the restoration degree reaches a specified value, or the state of the restored image is sequentially observed by the image output unit 104. When the observer determines that it is sufficient, the input operation unit 103 stops the restoration process and records the restored image on the processed image recording medium 105. Of course, if it is determined that the image recorded on the recording medium 101 is acceptable to be observed by the image output unit 104, the image can be recorded on the processed image recording medium 105 without performing the image restoration processing.

Here, the configuration of the image processing operation circuit 102 in the present embodiment will be described with reference to FIG. The image processing operation circuit 102 has a function of inputting a degraded image due to blurring or the like and outputting a restored image from which the cause of deterioration has been removed.
Of a degraded image by a blind deconvolution method such as an image restoration algorithm by the Fourier iteration method.

The image processing arithmetic circuit 102 includes an arithmetic circuit 600 for performing Fourier transform and four arithmetic operations, Fourier transform information of a blurred image, estimated restored image information updated each time a restoration process is performed, and an estimated point image as a deterioration factor. An arithmetic image storage memory 601 stores response function information, Fourier transform information of an estimated restored image, Fourier transform information of an estimated point image response function, and the like.

Here, the estimated point image response function information which is a deterioration factor is the same image information as the estimated restored image information, and each image information is two-dimensional light intensity information, and is equal to the number of horizontal pixels × the number of horizontal pixels.
This is general image information in which light intensity information is expressed in a matrix of the number of vertical pixels. Further, each Fourier transform information is a Fourier transform information of the image information, and is a general two-dimensional frequency information, in which phase and amplitude information is expressed in a matrix having the same number of pixels as the image information. This is frequency information.

The outline of the arithmetic processing of the arithmetic circuit 600 in FIG. 4 will be described with reference to FIG. The image restoration algorithm by the Fourier iteration method is a cyclic algorithm by iterative calculation. FIG. 5 shows an outline of the circulation algorithm. In the arithmetic circuit 600 shown in FIG.
Means 700 for performing a Fourier transform operation, means 701 for performing a reciprocal operation and an integration operation, means 702 for performing an inverse Fourier transform operation, and means 703 for performing a positive / negative judgment and an addition / integration operation.
It is composed of

The means 700 for performing a Fourier transform operation is a means for performing a Fourier transform on the estimated reconstructed image information in the restoration process or the estimated point image response function information which is a cause of deterioration. By performing Fourier transform on each image information, Fourier transform information can be obtained. In the outline of the calculation processing in FIG. 5, a circular algorithm is shown, so that the function of performing Fourier transform on the information of the input blurred image is not shown, but this function is also performed by the means 700 for performing Fourier transform calculation. Are stored in the operation image storage memory of FIG.

The means 701 for performing the reciprocal operation and the integration operation is
Fourier transform information of the estimated restored image, which is Fourier transformed by the means 700 for performing a Fourier transform operation, or
Reciprocal operation of the Fourier transform information of the estimated point image response function,
This is means for performing an integration (multiplication) operation with the Fourier transform information of the blurred image. Means 7 for performing this reciprocal operation and integration operation
01 performs the deconvolution processing shown in equation (3). Means 701 for performing reciprocal operation and integration operation
The Fourier transform information of the blurred image used in the processing of
The Fourier transform information of the estimated restored image or the Fourier transform information of the estimated point image response function, which is read from the operation image storage memory of FIG.
Is stored in the calculated image storage memory.

The means 702 for performing the inverse Fourier transform operation includes:
This is a means for performing an inverse Fourier transform of the Fourier transform information of the estimated restored image or the Fourier transform information of the estimated point image response function calculated by the means 701 for performing the reciprocal operation and the integration operation. Image information can be obtained by performing inverse Fourier transform on each piece of Fourier transform information. The means 700 for performing the Fourier transform operation and the means 702 for performing the inverse Fourier transform operation perform the same processing as the arithmetic processing, and the same arithmetic circuit can be used.

Means 703 for performing positive / negative judgment and addition integration operation
Is a means for giving a non-negative condition for each pixel of each image information as shown in FIG. 3, and calculates an arithmetic circuit for determining whether the value of each pixel is positive or negative, and calculates a sum of absolute values of negative pixels. It comprises an adding circuit, an integrating circuit for averaging the sum of absolute values of negative pixels, and an adding circuit for adding the average of the sum of absolute values of negative pixels to the values of all pixels.

The occurrence of a negative value in the light intensity information contradicts the physical phenomenon. The light intensity information indicates whether the image is bright or dark, and it is inconsistent with physical phenomena that a negative value, for example, a black hole that absorbs energy appears in the image information. . However, by giving the non-negative condition by the means 703 for performing the positive / negative judgment and the addition / integration calculation, it is possible to obtain image information having no contradiction as a physical phenomenon. Means 7 for performing positive / negative judgment and addition integration calculation
By transmitting the obtained image information (estimated restored image information or estimated point image response function information) to the means 700 for performing a Fourier transform operation, the cyclic algorithm functions. In this circulation algorithm, one restoration processing of the estimated restored image information or the estimated point image response function information is performed in one circulation, and the blind decombo shown in FIG. One iterative calculation of the image restoration processing by the lution method is performed.

Means 703 for performing positive / negative judgment and addition / integration calculation
The estimated restored image information or the estimated point image response function information updated with the operation result is stored in the operation image storage memory of FIG. The updated estimated restored image information is appropriately output to the restored image in FIG. 4 (the image output unit 1 in FIG. 1).
04) and can be observed by an observer. Until the output of this restored image is determined to be sufficient for the viewer,
The cyclic algorithm of FIG. 5 is iteratively calculated.

As described above in detail, in the digital camera having the image processing function according to the first embodiment, by using the degraded image restoration method by the blind deconvolution method as the image processing function, the conventional camera can be used. It eliminates the need for a sensor for detecting some physical quantity for the correction factor such as camera shake, which is a problem. In addition, since a mechanism for automatically detecting the amount of blur is not used, the correction can be normally performed along the subject image captured by the operator, and an image of an incorrect correction result is prevented from being recorded on a recording medium. can do.

The digital camera having the image processing function according to the first embodiment has a recording medium for recording both a raw image and an image after image processing for an image to be taken. If the blur amount is automatically measured and corrected as in, even if the amount of incomplete blur is measured, no further correction is possible, and correction that deviates extremely from the raw image Therefore, recording of an image of an incomplete correction result can be prevented.

Further, since the digital camera having the image processing function according to the first embodiment has the image processing function by the degraded image restoration method by the blind deconvolution method, the image restoration by the computer such as a personal computer is performed. No processing is required. Therefore, the processed image recording medium 1
By using a detachable recording medium 05 or the like, it is possible to directly output a good image in which camera shake and the like are restored to an external output device (not shown) such as a printer.

Next, the configuration of a digital camera having an image processing function according to the second embodiment will be described with reference to FIG. The basic configuration of a digital camera having an image processing function is the same as that of the first embodiment. The difference from the first embodiment is that the input operation unit 1 according to the first embodiment is different from the first embodiment.
On the other hand, in the second embodiment, an input operation unit 400 with a mode switching function is configured, and further, a point image response function recording medium 401 is configured.

The point image response function recording medium 401 is constituted by, for example, a flash memory, a hard disk or the like, like the recording medium 101 and the like. The point image response function recording medium 40
In a restoration process of a deteriorated image such as a camera shake by the blind deconvolution method described in the first embodiment, a point image response function serving as a deterioration factor, which is obtained simultaneously with a restored image, can be appropriately recorded and read. It is something.

The input operation unit 400 with the mode switching function comprises switches and a touch panel, like the input operation unit 103 in the first embodiment. The input operation unit 400 with the mode switching function is capable of appropriately recording and reading out the point image response function on the point image response function recording medium 401 together with the basic function of the input operation unit 103 in the first embodiment. Operation function.

Basic operation in the second embodiment
The effect is the same as that of the first embodiment. The difference from the first embodiment is that an input operation unit 400 with a mode switching function is configured instead of the input operation unit 103, and that a point image response function recording medium 401 is further configured. As described in the first embodiment, the restoration processing of a deteriorated image such as a camera shake by the blind deconvolution method is a method of estimating and separating F and G in Expression (2) from only the information of the deteriorated image. Therefore, when the restoration processing of the deteriorated image is completed, it is also possible to obtain information of the point image response function that is a deterioration factor together with the restored image that has not deteriorated.

In the second embodiment, a point image response function which is a cause of this deterioration is recorded on a point image response function recording medium 401, and when another degraded image is restored, it is actively used. By using this, it is possible to reduce the convergence processing time by iterative calculation. That is, the restoration processing of the deteriorated image by the blind deconvolution method is, for example,
It is performed by an iterative calculation as shown in FIG. In this iterative calculation, C = FG in Expression (2), where C is the Fourier transform of the degraded image, F is the Fourier transform of the original image, and G is the Fourier transform of the point image response function that is a degradation factor, Is known in the obtained degraded image. To start the iterative calculation, the initial value of F or G is estimated. Normally, F and G are completely unknown, so random numbers are used as initial values. Here, for example, the initial value is F of a random number. G for this initial value is obtained from the relationship of equation (2). The G is given a non-negative constraint condition as shown in FIG. Using the G approaching the correct answer, F is obtained from the relationship of equation (2).
Similar to G, non-negative constraint conditions are applied to this F to make it closer to the correct answer. By repeating this operation by iterative calculation, both F and G approach the correct answer.

In this iterative calculation, if the iterative calculation is started from the one whose initial value is close to the correct answer, the convergence time is also reduced. Therefore, for example, a point image response function obtained as a result of performing a restoration process on a deteriorated image when the degree of camera shake is large or small is recorded in the point image response function recording medium 401, and another deteriorated image is When restoring the
By selecting a mode with a large camera shake or a mode with a small camera shake by the input operation unit 400 with the mode switching function, the restoration processing time can be reduced.

In the digital camera having the image processing function according to the second embodiment, the deterioration factor represented by the formula (1) or (2) can be restored. Restoration processing is possible not only due to camera shake, but also for example, deterioration due to defocus of the lens, deterioration due to the flow of images when the subject is moving, deterioration due to fluctuations in the atmosphere (eg, heat haze), and the like. . Therefore, it is possible to add each mode for each such deterioration factor to the input operation unit 400 with the mode switching function.

The image processing function according to the second embodiment can of course be applied to the storage of a still image in a conventional digital camera or digital video camera.

As described in detail above, in the second embodiment, as in the first embodiment, a camera shake or the like is obtained by using a degraded image restoration method by a blind deconvolution method as an image processing function. The need for a sensor to detect some physical quantity,
Equipped with a recording medium for recording both the raw image and the image after image processing for the image to be captured. In addition to the fact that it has an image processing function using a degraded image restoration method by the blind deconvolution method, it eliminates the need for image restoration processing by a computer such as a personal computer. A point image response function recording medium 401 capable of appropriately recording and reading functions is provided, and when restoring another deteriorated image, an appropriate one is selected from the recorded point image response functions. Providing the input operation unit 400 with a mode switching function that enables the mode has an advantage such as that it is possible to shorten the time required for the degraded image restoration processing.

[0052]

As described above, the present invention provides:
By using a degraded image restoration method by a blind deconvolution method as an image processing function, a sensor for detecting any physical quantity for correction factors such as camera shake is not required, and a good image can be obtained with a simple configuration. It is possible to provide a digital camera having an image processing function. In addition, since a mechanism for automatically detecting the amount of blur is not used, the correction can be normally performed along the subject image captured by the operator, and an image of an incorrect correction result is prevented from being recorded on a recording medium. can do.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a digital camera having an image processing function according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an image restoration algorithm by a Fourier iteration method, which is the content of the image processing function according to the embodiment of the present invention.

FIG. 3 is a non-negative condition model diagram of an image restoration algorithm by a Fourier iteration method.

FIG. 4 is a configuration diagram of an image processing operation circuit according to the embodiment of the present invention.

FIG. 5 is a schematic diagram of a calculation process of the image processing calculation circuit according to the embodiment of the present invention.

FIG. 6 is a configuration diagram of a digital camera having an image processing function according to a second embodiment of the present invention.

FIG. 7 is a schematic configuration diagram of a method for correcting image deterioration due to camera shake or the like in a conventional digital camera.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 101 Storage medium (storage device) 102 Image information calculation circuit (image processing means) 103 Input operation part (operation means) 104 Image output part (display device)

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2H054 AA01 5B057 BA11 BA23 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CE03 CH08 5C022 AA13 AB55 AC00 AC01 AC31 AC69 5C053 FA08 FA30 HA40 KA04 KA22 KA24 KA30 KA30 LA01

Claims (4)

[Claims] An image processing means for correcting a captured image in a digital camera including a storage device for storing a captured image and a display device for displaying the image and having an image processing function for correcting the image. Operating means for providing correction information to the image processing means, wherein the operating means is configured to be operable from the outside, and based on a captured image displayed on the display device, the correction information is appropriately provided by the operating means. A digital camera having an image processing function, wherein an image is corrected by adding an image. 2. The digital camera having an image processing function according to claim 1, wherein a degraded image restoration method using a blind deconvolution method is used as the image processing function. 3. The storage device according to claim 1, wherein the storage device includes a pre-correction image storage unit that stores an image before correction, and a post-correction image storage unit that stores a post-correction image. Digital camera with image processing function. 4. The image according to claim 1, wherein the storage device includes an image correction information storage unit that stores a history of the image correction information, and the history can be used by the operation unit. Digital camera with processing function.