JP2004139574A - Image correction method, image correction device, and image correction program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce processing time on an image correction method, an image correction device, and an image correction program which detect and correct a prescribed defect about eyes in an image based on image data indicating the image. <P>SOLUTION: This system is provided with a pre-processing step (step S02, S04) which conducts a pre-processing that targets at least either the image indicated by the image data or the area on the image and narrows down what meets a prescribed condition indicating that the defect can exist; and a correction step (step S03, S07) which detects and corrects the defect for the result of the pre-processing step based on the image data. <P>COPYRIGHT: (C)2004,JPO

Description

The present invention relates to an image correction method, an image correction apparatus, and an image correction program for detecting and correcting defects such as red-eye, gold-eye, and blinking of eyes in an original image based on image data representing an image.

Conventionally, with the spread of digital still cameras, digitization in the field of photography is rapidly progressing. An advantage of such digitization is that the photographed image can be processed electronically to correct color and exposure defects, and the photographed image, referred to as photo retouching software, can be used as a personal computer. Software tools that modify the above are also becoming popular.

In recent years, there have been problems such as the red-eye phenomenon in which the pupil becomes red when flashing a person, the gold-eye phenomenon in which the eyes shine in gold, or the blinking state in which the person closed his eyes during shooting. In addition, a technique for correcting this defect has also been proposed. For example, Patent Document 1 describes a technique for detecting and correcting red eyes. Such defects greatly affect the quality of appearance, and the frequency of occurrence is high when shooting by a general photographer. Therefore, correction is particularly desired in the field of photography.
JP 2000-76427 A

However, when the detection and correction of defects according to the prior art are actually executed on a personal computer or the like, for example, when a method such as template matching is used for image data of 1500 × 1000 pixels, for example, about several seconds for one image It takes a lot of processing time. For example, assuming that such a defect detection / correction function according to the prior art is installed in a lab device that creates a photographic print from a photographic film at a store, the time required for the processing is too long and the application is impossible. is there.

Such a problem is not a problem that occurs only in the photographic field, but can occur in any field when the processing time is limited.

In view of the above circumstances, an object of the present invention is to provide an image correction method, an image correction apparatus, and an image correction program capable of reducing processing time.

The image correction method of the present invention that achieves the above object is an image correction method for detecting and correcting a specific defect related to an eye in an image based on image data representing the image.
A preprocessing step of performing preprocessing for narrowing down an image represented by image data and a predetermined condition indicating that the above-described defect may exist for at least one of an area on the image, and
What remains in the pre-processing process includes a correction process for detecting and correcting the above-described defect based on image data.

According to the image correction method of the present invention, an image or a region that is not likely to cause a defect is excluded from the correction target by pre-processing, and is narrowed down to only those that may cause a defect. Such pre-processing can be executed easily and in a short time based on a simple index or the like, so that the efficiency of the correction processing is improved as a whole and the processing time is reduced. In addition, in such pre-processing, a defect that does not clearly have a defect is excluded from the correction target, so that it is expected to reduce the erroneous detection of the defect.

Here, as a form of the image correction method of the present invention,
The image data represents a captured image, and attached with shooting information obtained at the time of shooting,
A mode in which the preprocessing step is a step of performing preprocessing based on shooting information attached to image data is a preferable mode.

The shooting information attached to the image data includes, for example, Exif information, flash light on / off information, shutter speed, aperture value, exposure (Ev value), subject distance, shooting mode, top and bottom information, shooting time, and camera model. Etc. may be employed. When such shooting information is used, for example, a red-eye phenomenon or a gold-eye phenomenon that is known to occur during flash shooting is performed when the on / off information of the flashlight device indicates off or when the shooting time is daytime. It is possible to easily eliminate those that cannot cause a problem, such as not occurring when the aperture value is small, and efficiently leave only those that may have a problem.

It is also preferable that the preprocessing process is a process of performing preprocessing based on scene analysis of an image represented by image data.

For scene analysis, for example, flash shooting detection, human face area extraction, blue sky detection, ground detection, and the like can be used. Flash shooting detection includes, for example, the degree of under, average density, dynamic range, and RGB values. Whether or not the flash photography is performed is estimated from the maximum value and the minimum value. By using such scene analysis, for example, it is possible to limit the correction target to a human face area. When the target area is limited by pre-processing, the processing time can be significantly reduced and mounting on lab equipment becomes easy.

In addition, as a form of the image correction method of the present invention, it is also preferable to further include a display process for displaying a result of the preprocessing by the preprocessing process.

* By displaying the result of pre-processing, it is possible to confirm which target is corrected before correction.

In addition, as a form of the image correction method of the present invention,
The method further includes a correction process for confirming the result of the preprocessing by the above preprocessing process and reselecting the one having the defect,
It is preferable that the correction process is a process of detecting and correcting the defect based on the image data with respect to what has been narrowed down through the preprocessing process and the correction process.

For specific defects, even if it is difficult to determine the presence or absence of defects by shooting information or scene analysis, an operator may be able to instantaneously determine an image that cannot occur. In such a case, the operator confirms the result of the preprocessing by the computer, corrects the error of the preprocessing, and further narrows down the target, thereby improving the accuracy of the entire image correction processing and further increasing the processing time. Can be reduced.

The image correction apparatus of the present invention that achieves the above-described object provides:
Based on image data representing an image, in an image correction apparatus that detects and corrects a specific defect related to eyes in the image,
A pre-processing unit that performs pre-processing for narrowing down an image represented by image data and a predetermined condition indicating that the above-described defect may exist for at least one of an area on the image, and
What remains in the pre-processing step is characterized by having a correction unit that detects and corrects the defect based on image data.

An image correction program of the present invention that achieves the above object is provided as follows:
In an image correction program for detecting and correcting a specific defect related to eyes in an image based on image data representing an image,
A pre-processing unit that performs pre-processing for narrowing down an image represented by image data and a predetermined condition indicating that the above-described defect may exist for at least one of an area on the image, and
An image correction program comprising: a correction unit that detects and corrects the above-described defect based on image data for what remains in the preprocessing process.

Note that the image correction device and the image correction program according to the present invention are only shown in the basic form here, but this is only for avoiding duplication, and the image correction device and the image correction program according to the present invention. Includes not only the above basic form but also various forms corresponding to the above-described forms of the image correction method.

Moreover, in the image correction apparatus of the present invention and the image correction program, the same names such as pre-processing units are given as component names constituting them, but in the case of an image correction program, It refers to software that performs such an action, and in the case of an image correction apparatus, it refers to software that includes hardware.

Further, the constituent elements such as the pre-processing unit constituting the image correction program of the present invention may be one in which the function of one constituent element is carried by one program component, and the function of one constituent element may be a plurality of functions. It may be carried by a program part, or the functions of a plurality of components may be carried by one program part. In addition, these components may execute such actions themselves, or may be executed by giving instructions to other programs and program components incorporated in the computer. Also good.

According to the image correction method, the image correction apparatus, and the image correction program of the present invention, the processing time can be reduced.

Hereinafter, embodiments of the present invention will be described.

Here, one embodiment of the image correction program of the present invention is incorporated in a personal computer, and one embodiment of the image correction method of the present invention is executed on the personal computer based on the image correction program. An example in which the computer operates as an embodiment of the image correction apparatus of the present invention will be described.

FIG. 1 is an external view of a personal computer to which an embodiment of the present invention is applied.

The personal computer 100 includes a main body 110 incorporating a CPU, a RAM memory, a hard disk, etc., a CRT display 120 that displays a screen on the phosphor screen 121 according to an instruction from the main body 110, and user instructions and characters in the personal computer 100. A keyboard 130 for inputting information and a mouse 140 for inputting an instruction corresponding to the position by designating an arbitrary position on the fluorescent screen 121 are provided.

The main body 110 further has a flexible disk loading port 111 and a CD-ROM loading port 112 into which a flexible disk and a CD-ROM are loaded in appearance, and inside the loaded flexible disk and CD. -A flexible disk drive and a CD-ROM drive for driving the ROM are also incorporated.

Here, an image correction program according to the present invention is stored in a CD-ROM, and this CD-ROM is loaded into the main body 110 from a CD-ROM loading slot 112 and is loaded into the CD-ROM by a CD-ROM drive. The stored image correction program is installed in the hard disk of the personal computer 100. When the image correction program installed in the hard disk of the personal computer 100 is started, the personal computer 100 operates as an embodiment of the image correction apparatus of the present invention, and an embodiment of the image correction method of the present invention. Execute.

In the present embodiment, image data of a photographed image taken with a digital still camera or the like is stored in a flexible disk or CD-ROM, and the flexible disk or CD-ROM is loaded into a personal computer 100 as an image correction apparatus. The image data is loaded into the hard disk. The captured image data is then subjected to red-eye correction and other image processing that corrects the black eye by detecting the red-eye phenomenon in which the flash emitted from the camera during so-called flash photography is reflected on the retina of the eye and the pupil glows red. Then, the processed image data is recorded on a flexible disk or hard disk.

FIG. 2 is a flowchart showing the first embodiment of the image correction method of the present invention.

In this image correction method, first, in the correction mode selection process (step S01), the operator confirms the “basic mode” in which correction processing (described later) is performed according to the result of preprocessing (described later) and the result of the preprocessing. Then, one of the “confirmation modes” in which the correction process is performed is selected. The “basic mode” includes a preprocessing process (step S02) and a correction process (step S03), and the “confirmation mode” includes a preprocessing process (step S04) and a correction process (step S04) similar to the “basic mode”. In addition to S07), a display process (step S05) and a correction process (step S06) are included.

The pre-processing process is a process of narrowing down conditions that satisfy a predetermined condition indicating that a defect may exist, and corresponds to an example of a pre-processing process in the image correction method of the present invention. Further, in the present embodiment, preprocessing for an image is performed in the preprocessing process, and in the preprocessing, the possibility of occurrence of a red-eye phenomenon is determined based on photographing information.

Also, the display process is a process of displaying the target image narrowed down by the preprocessing, and corresponds to an example of the display process in the image correction method of the present invention.

The correction process is a process in which the operator confirms the result of the preprocessing and narrows down the target image in which a defect exists. This correction process corresponds to an example of a correction process in the image correction method of the present invention.

Here, in the present invention, the correction mode selection process, the display process, and the correction process are not essential.

Further, the correction process is a process in which a defect is detected and corrected for the target image narrowed down in the preprocessing process and the correction process, and corresponds to an example of the correction process in the image correction method of the present invention. In the present embodiment, the red-eye phenomenon is detected and corrected in this correction process.

The detailed contents of each step will be described later.

FIG. 3 is a diagram showing a first embodiment of the image correction program of the present invention. Here, the image correction program 300 is stored in the image correction program storage medium 200.

As long as the image correction program storage medium 200 shown in FIG. 3 is a storage medium in which the image correction program 300 is stored, the type thereof is not limited. For example, when the image correction program 300 is stored in a CD-ROM. Indicates the CD-ROM, and when the image correction program 300 is loaded and stored in the hard disk device, indicates the hard disk device, or the image correction program 300 is downloaded to a flexible disk, DVD, CD-R or the like. Refers to the flexible disk.

The image correction program 300 is executed in the personal computer 100 shown in FIG. 1, and operates the personal computer 100 as an image correction apparatus that detects and corrects the red-eye phenomenon in the image. A display unit 320, a correction unit 330, and a correction unit 340.

The preprocessing unit 310, the display unit 320, the correction unit 330, and the correction unit 340 respectively include a preprocessing process (steps S02 and S04), a display process (step S05), and a correction process (step S06) illustrated in FIG. It also plays a role of executing the correction process (steps S03, S07). The preprocessing unit 310, the display unit 320, the correction unit 330, and the correction unit 340 correspond to examples of the preprocessing unit, the display unit, the correction unit, and the correction unit, respectively, in the image correction program of the present invention. .

The operation of each element of the image correction program 300 will be described later.

FIG. 4 is a functional block diagram of the first embodiment of the image correction apparatus of the present invention.

The image correction apparatus 400 is configured by installing and executing the image correction program 300 of FIG. 3 in the personal computer 100 shown in FIG.

The image correction apparatus 400 includes a preprocessing unit 410, a display unit 420, a correction unit 430, and a correction unit 440. The preprocessing unit 410, the display unit 420, the correction unit 430, and the correction unit 440 respectively constitute the image correction program 300 shown in FIG. 3, and the preprocessing unit 310, the display unit 320, the correction unit 330, and the correction unit 340, respectively. Corresponds to an example of each of the preprocessing unit, the display unit, the correction unit, and the correction unit in the image correction apparatus of the present invention. 4 is composed of a combination of the hardware of the personal computer 100 shown in FIG. 1 and an OS or application program executed on the personal computer, whereas the image correction program shown in FIG. These elements are different only in that they are constituted only by application programs.

Hereinafter, by describing each element of the image correction apparatus 400 shown in FIG. 4, each step of the flowchart shown in FIG. 2 and each element of the image correction program 300 shown in FIG.

First, the case where the basic mode is selected as the correction mode will be described.

When the operator selects “basic mode” on a correction mode selection screen (not shown), the correction mode information D11 is transmitted to the preprocessing unit 410. In the flowchart of FIG. 2, the process proceeds from step S1 to step S2.

Image data D10 representing a plurality of original images is input to the pre-processing unit 410, and based on shooting information attached to the image data D10, an original image that may have a red-eye phenomenon is generated in the image. This is narrowed down (step S2 in FIG. 2). Here, the image data D10 is obtained by a digital still camera. The image data D10 includes Exif information, on / off information of the flash light emitting device, shutter speed, aperture value, exposure (Ev value), subject distance, Shooting information such as shooting mode, top and bottom information, shooting time, camera model, etc. is attached. Then, the preprocessing unit 410 uses this shooting information to narrow down the original image to be corrected. For example, when the on / off information of the flashlight device indicates OFF, when the shutter speed is fast and the exposure (Ev value) is high, when the aperture value is small and the shooting time indicates daytime, or the subject distance If the distance exceeds the effective flash range, or if the shooting mode is landscape mode, there is no possibility that red-eye has occurred, so it is excluded from red-eye correction. On the other hand, if the camera model indicates a model that tends to cause red eyes and the on / off information of the flash light emitting device indicates ON, the camera is subjected to red eye correction. Such pre-processing in the pre-processing unit 410 can be executed at high speed with a simple determination routine or the like even if there are many determination items, so that the target can be efficiently narrowed down even if the processing time is short. it can. In addition, the preprocessing in the preprocessing unit 410 is a process whose purpose is to reduce waste in the subsequent correction process, and is not intended to reliably determine whether or not red eyes have occurred. For this reason, by using only the determination items for determining whether the red-eye phenomenon does not exist reliably or can exist even a little as the determination items, it is possible to sufficiently achieve the purpose and prevent a determination error. .

The result of the preprocessing in the preprocessing unit 410 is sent to the correction unit 440 together with the image data D10. Of the image data D10, image data representing a document image determined to have red eyes as a result of the preprocessing is subjected to correction processing for detecting and correcting red eyes (step S3 in FIG. 2). Since the content of the correction process itself is not the subject of the present invention, description thereof will be omitted. For example, a technique disclosed in Japanese Patent Laid-Open No. 2000-76427 is used. Since the target of the correction process is narrowed down by the preprocessing unit 410, the processing time in the correction unit 440 is shortened, and there is little false detection of the red-eye phenomenon.

The red-eye correction image obtained by performing the correction process by the correction unit 440 and the original image excluded from the correction process target by the pre-processing unit 410 are shown in FIG. The image data is sent to another image processing 500 such as color processing or sharpness processing executed in the personal computer 100.

This completes the explanation when “Basic mode” is selected, and then the explanation when “Confirm mode” is selected.

Even when the operator selects “confirmation mode” on the correction mode selection screen (not shown), the correction mode information D11 is transmitted to the preprocessing unit 410. In the flowchart of FIG. 2, the process proceeds from step S1 to step S4.

The pre-processing unit 410 narrows down an original image that may have a red-eye phenomenon in the image by the same process as step S02 in FIG. 2 (step S4 in FIG. 2).

The result of this preprocessing is displayed on the CRT display 120 shown in FIG. 1 by the display unit 420 (step S5 in FIG. 2).

FIG. 5 is a diagram showing a display example of a preprocessing result and an example of a correction screen.

5 (A) shows a fluorescent screen 121 of a CRT display, and all the original images 421 input to the preprocessing are reduced and displayed on the fluorescent screen 121. Further, a red-eye mark 422 is added to the original image 421 narrowed down by the preprocessing.

The operator confirms the original image 421 displayed on the CRT display 120 and corrects the result of the preprocessing (step S6 in FIG. 2).

When the right button of the mouse 140 shown in FIG. 1 is clicked on the original image 421, a correction pop-up screen 122 shown in Part (B) of FIG. 5 is displayed. The operator confirms the original image 421 to which the red-eye mark 422 is added, and selects “cancel red-eye” on the correction pop-up screen 122 when there is no possibility of the red-eye phenomenon at first glance. On the other hand, when the red-eye phenomenon has clearly occurred in the original image 421 to which the red-eye mark 422 is not added, “add red eye” is selected on the correction pop-up screen 122. The correction unit 430 illustrated in FIG. 4 excludes the original image for which “red-eye cancellation” has been selected by the operator from the target for red-eye correction, and adds the original image for which “red-eye addition” has been selected to the target for red-eye correction.

Here, another display example that is not employed in the present embodiment but may be employed in the present invention will be described.

FIG. 6 is a diagram showing another display example of the preprocessing result.

In the case of the display example shown in FIG. 6, only the original image 421 narrowed down by the preprocessing is displayed on the phosphor screen 121. For example, by setting the condition for narrowing down the original image in the pre-processing unit 410, all the original images that may have a red eye phenomenon are displayed, and the operator Only “red-eye cancellation” may be performed. Since the number of images to be confirmed is reduced by the preprocessing, such confirmation is expected to be easy.

The result of the preprocessing corrected through the correction unit 430 is sent to the correction unit 440 together with the image data D10. Of the image data D10, image data representing a document image that has been determined to have red eyes as a result of the preprocessing after correction is the same as the correction process in the “basic mode” (step S3 in FIG. 2). The correction process is performed in the correction process (step S7 in FIG. 2). Similar to the “basic mode”, the image data D22 representing the red-eye corrected image with the corrected red-eye is sent to the other image processing described above.

In addition to the description with reference to FIGS. 5 and 6, for example, all the original images 421 input in the pre-processing are displayed on the CRT display 120, and “red-eye addition” or “ It may be one in which “red-eye cancellation” is switched. The method of switching the addition and release of red eyes with one click has the advantage of easy operation.

Also, it is preferable that the pre-processing result, red-eye detection result, red-eye correction result, etc. are output to a file on the FD or printed on the back of the print paper. It can be easily checked later what processing steps have been performed.

As described above, the accuracy of the entire image correction process can be improved by performing the correction process after the operator corrects the result of the pre-process.

Next, a second embodiment different from the above-described embodiment will be described.

Also in the second embodiment, one embodiment of the image correction program of the present invention is incorporated in a personal computer, and one embodiment of the image correction method of the present invention is executed on the personal computer based on the image correction program. As a result, the personal computer operates as an embodiment of the image correction apparatus of the present invention.

Here, the second embodiment of the image correction method of the present invention and the second embodiment of the image correction program of the present invention will be described with reference to FIGS. 2 and 3 as they are.

In the second embodiment of the image correction method of the present invention, preprocessing is performed on a region on an image in a preprocessing process (steps S02 and S04) instead of the preprocessing process in the first embodiment. In the processing, the area is narrowed down based on the shooting information and the scene analysis. The area thus narrowed down is displayed in a display process (step S05) that replaces the display process in the first embodiment, and is corrected in a correction process (step S06) that replaces the correction process in the first embodiment. The

Further, in the second embodiment of the image correction program of the present invention, instead of the preprocessing unit, the display unit, and the correction unit in the first embodiment, the above-described preprocessing process for narrowing down the target area, the narrowed area Are provided with a pre-processing unit 310, a display unit 320, and a correction unit 330 corresponding to the display process for displaying the image and the correction process for correcting the narrowed area.

FIG. 7 is a functional block diagram showing a second embodiment of the image correction apparatus of the present invention.

The image correction apparatus 401 is configured by installing and executing the image correction program of the second embodiment on the personal computer 100 shown in FIG.

The image correction apparatus 401 includes a preprocessing unit 450, a display unit 460, a correction unit 470, and a correction unit 440. The preprocessing unit 450, the display unit 460, the correction unit 470, and the correction unit 440 correspond to examples of the preprocessing unit, the display unit, the correction unit, and the correction unit in the image correction apparatus of the present invention. 7 is exactly the same as the correction unit 440 shown in FIG.

Hereinafter, each element of the image correction apparatus 401 illustrated in FIG. 7 will be described.

The pre-processing unit 450 receives correction mode information D11 selected by the operator and image data D20 representing the original image, and top and bottom information D21 attached to the image data. Based on the image data D20 and the top and bottom information D21, the top and bottom detection unit 441 detects the top and bottom of the image. Further, the face detection unit 442 detects a face area in the original image by scene analysis based on the image data D20. Then, based on the top and bottom information detected by the top and bottom detection unit 441 and the face region detected by the face detection unit 442, the upper half half identification unit 443 identifies a region corresponding to the upper half of the face. The By specifying the region thus identified as a target region for red-eye correction, the target region for red-eye correction is significantly narrowed down. Note that an image or the like in which no face is detected by the pre-processing unit 440 is sent to other image processing such as color processing.

The area narrowed down in this way is displayed on the CRT display 120 shown in FIG.

FIG. 8 is a diagram illustrating an example of an original image (A), a detected face region (B), and a display example (C) of a preprocessing result in the second embodiment.

The original image 610 shown in part (A) of FIG. 8 is an example of the original image represented by the image data D20 shown in FIG. 7. The original image 610 shown in part (A) includes a person image 611. Yes.

Further, in part (B) of FIG. 8, a face region 620 detected by scene analysis based on the original image 610 shown in part (A) is indicated by hatching.

Further, in the part (C) of FIG. 8, the upper half area 630 of the face area 620 shown in the part (B) is indicated by a division mark 631. The image shown in the part (C) is displayed on the CRT display by the display unit. The operator can confirm whether or not the preprocessing has been performed accurately by viewing the image displayed in this way.

When the position of the division mark 631 in the displayed image is deviated at a glance, the operator moves the division mark 631 to a position where the red-eye phenomenon occurs using the mouse 140 shown in FIG. At this time, the correction unit 470 corrects the target area for red-eye correction in accordance with an instruction from the operator.

The image data of the original image displayed by the display unit 450 after the region is narrowed down by the preprocessing unit 440 shown in FIG. 7 is sent to the correction unit 440 after correcting the target region for red-eye correction through the correction unit 470. Thus, the above-described red-eye correction is performed on the target area. Since the target area is greatly narrowed down by the pre-processing unit 450, the processing time for red-eye correction is greatly reduced.

As a result of red-eye correction by the correction unit 440, image data D22 representing a red-eye corrected image with red-eye corrected is obtained, and this image data D22 is sent to the other image processing described above, as in the first embodiment.

In the above description, the example in which the face extraction process and the top / bottom determination process are performed independently for red-eye detection has been described, but these processes are performed as part of other processes such as a setup process, for example. It is desirable that the results be reused.

In the above description, red eyes are exemplified as eye defects. However, the eye defects referred to in the present invention may be defects other than red eyes, such as gold eyes and blinking eyes.

In the above description, face detection and top-and-bottom detection are shown as examples of scene analysis. However, in the present invention, detection of flash photography is adopted as scene analysis, and images that may have been obtained by flash photography. May be a correction target.

In the above description, the image correction program stored in the storage medium is exemplified, but the image correction program of the present invention may be transmitted / received over a communication network or the like.

In the above description, an image correction apparatus that captures image data stored in a storage medium is illustrated. However, the image correction apparatus of the present invention may capture image data from a digital still camera using a USB or the like. Alternatively, the image data may be captured via a communication network or the like.

Further, the image correction apparatus of the present invention may further include means for switching ON / OFF of preprocessing.

Further, in the above description, an example in which image data after image processing is recorded on a flexible disk or hard disk is shown. However, in the present invention, image data after image processing is stored in other formats such as CD-R, DVD, MO, etc. The image data may be output to a storage medium of the above type, the image data after image processing may be output to a photo printer or the like, or the image data after image processing may be output via a communication network or the like. It may be output.

In the above description, an example is shown in which other image processing is performed on the personal computer after the image correction according to the embodiment of the present invention. It may be separated from the image processing and executed independently by a dedicated personal computer or the like.

1 is an external view of a personal computer to which an embodiment of the present invention is applied. It is a flowchart which shows 1st Embodiment of the image correction method. It is a figure which shows 1st Embodiment of the image correction program of this invention. 1 is a functional block diagram of a first embodiment of an image correction apparatus of the present invention. It is a figure which shows the example of a display example of a pre-processing result, and the screen for the correction. It is a figure which shows another example of a display of a pre-processing result. It is a functional block diagram which shows 2nd Embodiment of the image correction apparatus of this invention. It is a figure which shows the example of an original image in 2nd Embodiment (A), the area | region (B) of the detected face, and the example of a display (C) of a pre-processing result.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Personal computer 110 Main body part 111 Flexible disk loading slot 112 CD-ROM loading slot 120 CRT display 121 Phosphor screen 130 Keyboard 140 Mouse 200 Image correction program storage medium 300 Image correction program 310 Preprocessing part 320 Display part 330 Correction part 340 Correction part 400, 401 Image correction device 410, 450 Pre-processing unit 421 Original image 422 Red-eye mark 441 Top-and-bottom detection unit 442 Face detection unit 443 Face upper half identification unit 420, 460 Display unit 430 Correction unit 440 Correction unit 500 Other image processing 610 Original Image 611 Human image 620 Face area 630 Upper half area D10, D20 Image data representing original image D11 Correction mode information D21 Top and bottom information D22 Red-eye correction image To image data

Claims (7)

In an image correction method for detecting and correcting a specific defect related to eyes in an image based on image data representing an image,
A preprocessing step of performing preprocessing for narrowing down an image represented by image data and a predetermined condition indicating that the defect may exist for at least one of an area on the image, and
An image correction method comprising: a correction process for detecting and correcting the defect on the basis of image data for those narrowed down by the preprocessing. The image data represents a captured image, and attached with shooting information obtained at the time of shooting,
2. The image correction method according to claim 1, wherein the preprocessing step is a step of performing the preprocessing based on photographing information attached to the image data. The image correction method according to claim 1, wherein the preprocessing step is a step of performing the preprocessing based on a scene analysis of an image represented by the image data. The image correction method according to claim 1, further comprising a display step of displaying a result of the preprocessing by the preprocessing step. Confirming the result of the pre-processing by the pre-processing process, further comprising a correction process for reselecting the defect is present,
2. The image correction according to claim 1, wherein the correction process is a process of detecting and correcting the defect on the basis of image data with respect to what has been narrowed down through the preprocessing process and the correction process. Method. Based on image data representing an image, in an image correction apparatus that detects and corrects a specific defect related to eyes in the image,
A pre-processing unit that performs pre-processing for narrowing down an image represented by image data and a predetermined condition indicating that the defect may exist for at least one of an area on the image, and
An image correction apparatus comprising: a correction unit that detects and corrects the defect based on image data for what remains in the preprocessing process. In an image correction program for detecting and correcting a specific defect related to eyes in an image based on image data representing an image,
A pre-processing unit that performs pre-processing for narrowing down an image represented by image data and a predetermined condition indicating that the defect may exist for at least one of an area on the image, and
An image correction program comprising: a correction unit that detects and corrects the defect based on image data for what remains in the preprocessing process.

Images