JP2014134996A - Correction quality checking device, correction quality checking method, program, and correction quality checking system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a correction quality checking device for efficiently checking correction quality after automatic correction processing on image data.SOLUTION: A control unit 11 of a server 3 causes image feature calculation means 21 to calculate image feature information 151 and 161 respectively for object image data 51, 61 (step S102). The image feature information 151 before automatic correction processing is compared with image feature information 131, and a flag is set in the image feature information 131 which is determined to have similarity among the image feature information 131 (step S103). When a flag is present in the image feature information 131, image feature information after correction processing 161 corresponding to the image feature information 151 of interest is extracted (S109). In after-processing image determination means 23, the image feature information 161 is compared with image feature information 141, and a flag is set in the image feature information 141 which is determined to have similarity among the image feature information 141 (step S110).

Description

  The present invention relates to a correction quality check apparatus and the like for performing a correction quality check after image data automatic correction processing more efficiently.

At present, DTP (Desk Top Publishing) is generally used as a print data creation method in printing plate making, and an image processing operator creates print data using a graphic application, a photo retouch application, and a typesetting application. Among them, software for handling image data is software classified as a photo retouch application, and an image processing operator performs image color correction that is more preferable or suitable for printing depending on the application.
However, in recent years, with the realization of faster delivery to customers or the development of automatic image color correction technology for image data, correction processing using automatic image correction technology has been performed instead of manual correction by image processing operators. The number of cases is increasing. Various automatic image color correction techniques have been developed (Patent Documents 1, 2, etc.).

Patent Document 1 discloses an invention such as an image processing apparatus that automatically determines a correction amount of image correction for target image data, and further finely corrects the correction amount manually and applies it to the target image data. .
The image processing apparatus described in Patent Document 2 applies image processing that is the same as the reference image to the target image data, and determines whether the difference between the two images exceeds a predetermined value. Inventions such as devices are disclosed.

JP 2005-208884 A JP 2012-063946 A

However, in the above prior art, it is necessary to determine a correction value for each image to which correction processing is to be applied, or the automatic image correction processing performed is not appropriate. The process of using and working will occur.
Also, at present, it is difficult for automatic image color correction technology to become a “complete operator replacement”. The reason is that although manual correction by the image processing operator has some common rules, the ratio of sensory elements of the image processing operator is strong, and it is not easy to faithfully reproduce it by automatic correction. In some cases, automatic correction may cause erroneous correction (obvious correction error, correction quality that does not conform to the intention of the image processing operator or customer). In such a case, an image processing operator reconfirms the correction result by using a photo retouching application for each image that has undergone automatic correction processing. Therefore, there is a problem that even if the automatic image correction processing device is used, the time required for the image correction processing is not shortened.

  The present invention has been made in view of such problems, and an object of the present invention is to provide a correction quality check apparatus and the like that more efficiently perform a correction quality check after automatic correction processing of image data.

In order to achieve the above-described object, the first invention provides, for a plurality of image data, image feature information before color correction of the image data and an image after color correction in which the image data is appropriately color corrected. Storage means for storing feature information in association with each other, image data input means for inputting target image data, and corrected image data obtained by automatically color correcting the target image data, the input target image data and the correction Image feature calculating means for calculating image feature information for each of the image data, and first determining whether image feature information before color correction similar to the image feature information of the target image data exists And the image feature information after color correction corresponding to the similar image feature information before color correction and the image feature information of the corrected image data. When it is determined that the second determination unit and the first determination unit do not exist, or when the second determination unit determines that they are not similar, the corrected image And an output means for outputting information indicating that automatic color correction of data is not appropriate.
According to the first invention, it is possible to determine the validity of the result of the automatic color correction using the processing result that the operator has appropriately corrected the color in the past as a reference. It is not necessary to confirm each processing image one by one, and the working efficiency is greatly improved.

The image feature information of the first invention preferably includes at least one piece of information indicating the feature of the image data, such as histogram information, black point information, white point information, contrast information, or spatial frequency information of the image data.
Accordingly, it is possible to determine whether or not automatic color correction of the corrected image data is appropriate using one or more pieces of image feature information.

It is desirable that the first determining means and the second determining means of the first invention determine a similarity by providing a threshold for the image feature information.
Thereby, the validity of the automatic color correction of the corrected image data can be given a range.

The image feature information after color correction stored in the storage means according to the first aspect of the invention is preferably information of a result that is recognized that appropriate color correction has been performed on the image data.
As a result, the image feature information of the image data with appropriate color correction and appropriate color correction can be used for determining whether or not the automatic color correction of the corrected image data is appropriate.

In the correction quality check device according to the first aspect of the present invention, a plurality of pairs of the target image data and the corrected image data are input by the image data input means, and information indicating that automatic color correction is not appropriate is output. It is desirable to further include display means for displaying image data or a list of the corrected image data.
As a result, it is possible to display a list of target image data or corrected image data for which automatic color correction is not appropriate.

In a second aspect of the present invention, image feature information before color correction of the image data and image feature information after color correction obtained by appropriately color-correcting the image data are stored in association with each other for a plurality of image data. A computer having storage means for inputting the target image data and corrected image data obtained by automatically color correcting the target image data; and each of the input target image data and the corrected image data. An image feature calculating step for calculating image feature information, and a first determination step for determining whether or not the image feature information before color correction similar to the image feature information of the target image data exists; If it is determined that it exists, the image feature information after color correction corresponding to the similar image feature information before color correction and the image feature information of the corrected image data Are determined in the second determination step and the first determination means, or when the second determination means determines that they are not similar, And an output step of outputting information indicating that automatic color correction of the corrected image data is not appropriate.
According to the second invention, it is possible to determine the validity of the result of the automatic color correction using the processing result that the operator has appropriately corrected the color in the past as a reference. It is not necessary to confirm each processing image one by one, and the working efficiency is greatly improved.

In a third aspect of the invention, the computer associates, for a plurality of image data, image feature information before color correction of the image data and image feature information after color correction obtained by appropriately color-correcting the image data. Storage means for storing image data, image data input means for inputting target image data and corrected image data obtained by automatically color correcting the target image data, an image for each of the input target image data and the corrected image data. Image feature calculation means for calculating feature information, first determination means for determining whether image feature information before color correction similar to the image feature information of the target image data exists, determined to exist In this case, it is determined whether the image feature information after color correction corresponding to the similar image feature information before color correction is similar to the image feature information of the corrected image data. If the second determination means, the first determination means determines that they do not exist, or if the second determination means determines that they are not similar, the automatic color correction of the corrected image data is not appropriate. This is a correction quality check program for functioning as output means for outputting information to that effect.
By installing the third invention in a general-purpose computer, the correction quality check device of the first invention can be obtained and the correction quality check method of the second invention can be executed.

  A fourth invention is a correction quality check system in which an operation terminal and a server are connected via a network, the operation terminal including target image data, corrected image data obtained by automatically color correcting the target image data, Input means for transmitting to the server, and display means for displaying the output of the server, the server for image data before color correction of the image data for a plurality of image data Storage means for storing information in association with image feature information after color correction in which the image data is appropriately color-corrected, means for receiving the target image data, and the corrected image data; Image feature calculation means for calculating image feature information for each of the target image data and the corrected image data, and the color correction similar to the image feature information of the target image data A first determination unit that determines whether or not the image feature information is present, and if it is determined that the image feature information is present, the image feature information after the color correction corresponding to the similar image feature information before the color correction, A second determination unit that determines whether or not the image feature information of the corrected image data is similar to the second determination unit; and the second determination unit that is determined not to exist in the first determination unit And output means for outputting information indicating that automatic color correction of the corrected image data is not appropriate when it is determined that they are not similar to each other in the correction quality check system.

  The correction quality check after the automatic correction processing of the image data can be performed more efficiently by the correction quality check device of the present invention.

The figure which shows the outline | summary of the correction | amendment quality check system 1 which concerns on embodiment of this invention. Hardware configuration diagram of information processing apparatus for realizing server 3 and operation terminal 5 Block diagram showing the functional outline of the server 3 The figure which shows the outline | summary of the data currently hold | maintained in the reference database in the memory | storage part 12. Diagram showing examples of image histogram information and multidimensional approximation function Diagram showing the outline of target image data A flowchart (1) showing an outline of processing of the control unit 11 of the server 3 Flow chart (2) showing an outline of processing of the control unit 11 of the server 3 The figure which shows the image feature information of object image data The figure which shows the outline of comparison of image histogram information The figure which shows the example of display of the judgment result

[1. Outline of correction quality check system 1]
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the accompanying drawings, the same reference numerals are given to components having substantially the same functional configuration, and redundant description is omitted.

  First, the outline of the correction quality check system 1 according to the embodiment of the present invention will be described with reference to FIG. The correction quality check system 1 uses an image group subjected to automatic correction processing as a check target, and an image (hereinafter referred to as a reference image) that has been appropriately corrected by an image processing operator (hereinafter referred to as an operator) as a reference image. ) Group to check the correction quality of a group of images to be checked (hereinafter referred to as target images).

  As shown in FIG. 1, the correction quality check system 1 is configured by connecting a server 3 (correction quality check device) and an operation terminal 5 via a network 7. The operation terminal 5 is a terminal device that can access the server 3 via a network 7 such as a personal computer.

In the correction quality check system 1, the server 3 receives user operation and instruction information from the operation terminal 5, and displays information on a screen for receiving operation and instruction from the user, and a check result. Is transmitted to the operation terminal 5.
The operation terminal 5 receives and displays screen information from the server 3, and transmits information input by the user on the displayed screen to the server 3 via the network 7. Note that the method of connecting the operation terminal 5 to the network 7 may be wireless or wired.

  FIG. 2 is a hardware configuration diagram of an information processing apparatus that realizes the server 3 and the operation terminal 5. Note that the hardware configuration in FIG. 2 is an example, and various configurations can be adopted depending on the application and purpose. As shown in FIG. 2, the information processing apparatus includes a control unit 11, a storage unit 12, an input unit 13, a display unit 14, a communication unit 15, and the like connected via a bus 16.

  The control unit 11 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The CPU calls and executes a program stored in the storage unit 12, ROM, recording medium, or the like to a work memory area on the RAM, drives and controls each device connected via the bus 16, and controls the server 3 and the operation terminal. The process which 5 mentioned later performs by 5 is implement | achieved. The ROM is a non-volatile memory and permanently stores programs, data, and the like. The RAM is a volatile memory, and temporarily stores a program, data, and the like loaded from the storage unit 12, ROM, recording medium, and the like, and includes a work area used by the control unit 11 to perform various processes.

  The storage unit 12 is a flash memory, a hard disk, or the like, and stores a program executed by the control unit 11, data necessary for program execution, and the like. Each of these program codes is read by the CPU of the control unit 11 as necessary, transferred to the RAM, and executed as various means. In addition to the OS (Operating System) that causes the control unit 11 of the server 3 to function and the correction quality check program, the storage unit 12 stores data related to the reference image to be compared with the check target image.

The input unit 13 inputs data and includes an input device such as a touch panel and a keyboard. An operation instruction, an operation instruction, data input, and the like can be performed via the input unit 13.
The display unit 14 includes a logic circuit or the like (video adapter or the like) for realizing a video function of a computer in cooperation with a display device such as a liquid crystal panel. The input unit 13 and the display unit 14 may be integrated like a touch panel display.

The communication unit 15 is a communication interface that mediates communication using a network, and communicates with other information processing apparatuses and the like via the network. The network may be wired or wireless.
The bus 16 is a path that mediates transmission / reception of control signals, data signals, and the like between the devices.

[1-1. Overview of server 3 functions]
FIG. 3 is a block diagram showing an outline of functions of the server 3. The server 3 includes an image feature calculation unit 21, a pre-processing image determination unit 22, a post-processing image determination unit 23, and a determination result output unit 24. These means are implement | achieved when the control part 11 of the server 3 reads the program (not shown) memorize | stored in the memory | storage part 12 to RAM, and executes it.

The image feature calculation means 21 calculates image features for all of the image data before automatic correction processing and the image data after automatic correction processing input from the operation terminal 5 by the user.
When the pre-process image determination unit 22 compares the image feature information of the target image data before the automatic correction process with the image feature information of the reference image data before the correction process, and determines that there is an item of similar image feature information In addition, a flag is set in the item of the image feature information of the target image data.

When the post-processing image determination unit 23 compares the image feature information of the target image data after the automatic correction processing with the image feature information of the reference image data after the correction processing, and determines that there is an item of similar image feature information In addition, a flag is set for the item of the image feature information of the target image data.
The determination result output unit 24 displays the result of the correction quality check on the display unit 14.

[1-2. Functions of two determination means]
By the way, when the correction process is performed on the image, the image feature information of the image changes before and after the correction process. In many cases, the same correction processing is performed on two images having image features similar to each other. In such a case, the image features of the two images after the correction processing are also similar to each other. Therefore, if the two image features are similar before and after the correction process, it can be said that the correction processes applied to the two images are equivalent.
Therefore, the inventors of the present invention applied this to the correction quality check system 1 of the present embodiment. In particular, the pre-processing image determination unit 22 is provided with a function of comparing two image features before correction processing to determine similarity, and the post-processing image determination unit 23 compares two image features after correction processing. The function to judge similarity was given. Then, using these two similarity determination results, correction processing (automatic correction processing and correction processing by an image operator in this embodiment) applied to each of two images (the target image and the reference image in the present embodiment). Can be determined by the correction quality check system 1 of the present embodiment, and the correction quality check can be performed more efficiently by specific means and methods to be described later. I was able to do it.

[2. Preparation of reference image data]
The correction quality check system 1 according to the present embodiment is a mechanism for referring to the manual image correction result of an operator such as an operator as a correct correction and determining the validity of the automatic image correction process. It is assumed that a reference is held as a database (hereinafter referred to as a reference database).
The reference database includes reference image data before correction processing by an operator (hereinafter referred to as reference image data before correction processing) and reference image data after correction processing by an operator (hereinafter referred to as reference image data). (Referred to as reference image data after correction processing). Further, image feature information (hereinafter referred to as image feature information) is calculated from these reference image data in advance, and image feature information about the reference image data before the correction process and an image about the reference image data after the correction process. Keep it as feature information. Each reference image and each image feature information are associated with each other by an ID, a file name, or the like. Although the number of image feature information records stored in the reference database is arbitrary, it is expected that determination accuracy will be improved by arranging a large number of all image correction patterns.

FIG. 4 shows an outline of reference image data held in the reference database in the storage unit 12.
The reference image is an image that has been appropriately corrected by the operator, and is a reference image when determining the correction quality of the target image. A plurality of reference images are prepared. In FIG. 4, reference image data 31 before correction processing and reference image data 41 after correction processing are prepared for each of n reference images. Furthermore, image feature information 131 is attached to each of the n reference image data 31 before the correction processing. Similarly, image feature information 141 is attached to each of the n reference image data 41 after the correction processing. Note that the reference image data 32 is the reference image data before the n-th correction process corresponding to the n-th reference image, and the reference image data 42 is after the n-th correction process corresponding to the n-th reference image. Reference image data and image feature information 132 are image feature information corresponding to reference image data 32, and image feature information 142 is image feature information corresponding to reference image data 42.

[3. Image feature information]
It is desirable that the image feature information related to the reference image data has a large number of image feature information items and can be arbitrarily selected from these items when compared with image feature information of a target image described later. In FIG. 4, the image feature information 131 and 141 includes image histogram information, black point information, white point information, contrast information, and spatial frequency information. The image feature information will be described below.

[3-1. Image histogram information]
The image histogram information of the reference image data includes a multidimensional approximation function indicating a histogram shape, and the multidimensional approximation function is calculated in advance from the value of each class. As a method for calculating the approximate function, for example, the nonlinear least square method is a representative calculation method.
Further, without using the approximate function, for example, the histogram value may be recalculated by using a moving average method using adjacent series values of the histogram, and the histogram information excluding the protruding series value may be recalculated. In order to ensure the validity of the similarity determination, it is desirable to calculate a rough histogram shape, and the method is not limited to the above method.

FIG. 5 shows an example of image histogram information and a multidimensional approximation function. 5A shows image histogram information and multidimensional approximation function curve 133 of reference image data 31 before correction processing, and FIG. 5B shows image histogram information and multidimensional approximation function of reference image data 41 after correction processing. Curve 143.
A threshold value is given to each of the curves 133 and 143 of the multidimensional approximation function. The upper limit value of the threshold value is represented by broken lines 134 and 144, and the lower limit value of the threshold value is represented by broken lines 135 and 145. The threshold value is a specific determination criterion when compared with image feature information of target image data described later. These thresholds may be set in advance when calculating a multidimensional approximation function, or the user may input arbitrary values when comparing with image feature information of target image data described later.

[3-2. Black point information, white point information]
The black point information is the minimum luminance value in the image data, and the white point information is the maximum luminance value. The reference image data 31 before correction processing and the reference image data 41 after correction processing are calculated and calculated. As a representative calculation method of luminance value, luminance value: Lum (x, y) = (R element + G element + B element) / 3
Minimum luminance value: Lum_min = MIN (Lum (x, y))
Maximum luminance value: Lum_max = MAX (Lum (x, y))
Here, (x, y) is the coordinates of the pixel.
The method is not particularly limited.
A threshold is given to each of the calculated minimum luminance value and maximum luminance value. The threshold value is a specific determination criterion when compared with image feature information of target image data described later. These threshold values are set in advance when calculating the minimum luminance value and the maximum luminance value, or the user inputs arbitrary values when comparing with image feature information of target image data described later. Also good.

[3-3. Contrast information]
The contrast information is a contrast value of the image data, and is obtained by calculating the reference image data 31 before the correction process and the reference image data 41 after the correction process. As a representative calculation method of the luminance value, the above-described minimum luminance value Lum_min and maximum luminance value Lum_max are calculated,
Contrast value: C
= (Lum_max-Lum_min) / (Lum_max + Lum_min)
The method is not particularly limited.

  A threshold is given to the calculated contrast value. The threshold value is a specific determination criterion when compared with image feature information of target image data described later. These threshold values may be set in advance when calculating the contrast value, or the user may input arbitrary values when comparing with image feature information of target image data described later.

[3-4. Spatial frequency information]
Spatial frequency information is obtained by calculating a spatial frequency distribution from image data by two-dimensional discrete Fourier transform, and is calculated and obtained for reference image data 31 before correction processing and reference image data 41 after correction processing.

  Where u = 0,. . . , M−1, v = 0,. . . , N−1, M is the number of pixels in the x direction of the image data, and N is the number of pixels in the y direction of the image data. f (x, y) is a two-dimensional signal (for example, image data), and is an x column y row component of f. F (u, v) is a two-dimensional frequency spectrum of f (x, y), u is the frequency of the x component, and v is the frequency of the y component.

  The above image feature information is prepared for all reference image data 31 before correction processing and reference image data 41 after correction processing, and is used for comparison with image feature information of target image data described below.

[4. Correction quality check]
Next, an outline of processing of the control unit 11 of the server 3 will be described. Prior to this process, the user prepares target image data to be subjected to the correction quality check, and inputs it from the input unit 13 of the server 3 or the operation terminal 5. FIG. 6 shows an outline of the target image data. On the left side of FIG. 6 (a), target image data 51 corresponding to each of the k target images and the folder 50 for storing them are displayed on the right side of FIG. 6 (a). Target image data 61 after processing and a folder 60 for storing them are prepared.

The outline of the process of the control unit 11 of the server 3 will be described with reference to the flowcharts of FIGS. 7 and 8 are flowcharts showing an outline of processing of the control unit 11 of the server 3.
First, the control unit 11 of the server 3 guides the display unit 14 to display both the target image data 51 before the automatic correction process and the target image data 61 after the automatic correction process (see FIG. 6 (b)) is displayed (step S101). The user of the correction quality check system 1 inputs a folder 50 in which target image data 51 before automatic correction processing is stored and a folder 60 in which target image data 61 after automatic correction processing is stored in accordance with the guidance screen.

  Next, the control unit 11 of the server 3 calculates image feature information 151 and 161 (see FIG. 9) for the input target image data 51 and 61 in the image feature calculation unit 21 (step S102), respectively. Are stored in folders 50 and 60. The image feature information is, for example, at least one piece of image feature information such as luminance (brightness), histogram information of each color of RGB, black point value, white point value, contrast value, spatial frequency characteristics, etc. It is desirable that items can be selected. Each target image and each image feature information are associated with each other by an ID, a file name, or the like.

In FIG. 6A, the target image data 52 is the target image data before the kth automatic correction process corresponding to the kth target image, and the target image data 62 is the kth target image corresponding to the kth target image. 9, the image feature information 152 is image feature information corresponding to the target image data 52, and the image feature information 162 is image feature information corresponding to the target image data 62.
The image feature information 151 and 161 corresponding to the target image data 51 and 61 is [3. It is calculated in the same manner as described in [Image Feature Information], but no threshold is given. Note that the image feature information 151 and 161 may be calculated before step S101 and stored in advance in the folders 50 and 60 as image feature information.

  Next, the control unit 11 of the server 3 uses the pre-processing image determination unit 22 to determine each determination item of the image feature information 151 (1 to k pieces) selected by the user corresponding to each of the k target image data 51, and The determination items of the image feature information 151 are extracted from the image feature information 131 (1 to n pieces) corresponding to the n pieces of reference image data 31 before correction processing stored in the reference database of the storage unit 12 and compared.

  Here, an example in which image histogram information is compared among items of image feature information will be described. The image histogram information calculated in step S102 and the curve 153 of the multidimensional approximation function (see the right of FIG. 9A) are the image histogram information (see FIG. 5A) of the image feature information 131 in step S103. ).

  Specifically, the multidimensional approximation function curve 153 included in the image histogram information of the image feature information 151 falls within the upper limit 134 and the lower limit 135 of the multidimensional approximation function threshold included in the image histogram information of the image feature information 131. Whether or not similarity is determined (see FIG. 10A). If the multidimensional approximate function curve 153 falls between the upper limit 134 and the lower limit 135, the image histogram information of the image feature information 151 and the image histogram information of the image feature information 131 are determined to be similar to each other. As described above, when the image feature information 151 is similar to the image feature information 131, a flag is set on the compared image feature information 131.

Note that, when similarity is determined for other items of image feature information, the similarity is determined based on whether the image feature information is within a threshold. In the case of spatial frequency information, a correlation coefficient is obtained between the spatial frequency distribution of the target image data 51 and the spatial frequency distribution of the reference image data 31, and the correlation coefficient is not less than the correlation threshold value set arbitrarily. It is determined that there is similarity, and it is determined that there is no similarity in the following cases.
When the user of the correction quality check system 1 selects a plurality of items of image feature information to be calculated, it is determined that there is similarity when the selected item is within a threshold value, and there is an item that does not satisfy even one item. It is determined that there is no similarity.

  In step S103, the process of step S103 is performed until the i-th image feature information 151 is compared with all the image feature information 131 (step S105). When the comparison with the n-th image feature information 132 is completed (Yes in step S104), the control unit 11 of the server 3 determines whether or not a flag exists in the image feature information 131 (step S106).

  When there is no flag in the image feature information 131 (No in step S106), there is a possibility that the corresponding target image data 51 has not obtained an optimum automatic correction processing result by the determination result output unit 24 of the server 3. The image data 81 (see FIG. 11) is displayed on the display unit 14 (step S107). Subsequently, the control unit 11 of the server 3 performs step S103 on the image feature information 151 corresponding to the i + 1th target image data 51. The same processing is performed (step S108).

  When there is a flag in the image feature information 131 (Yes in step S106), the control unit 11 of the server 3 extracts the image feature information 161 after the automatic correction process associated with the image feature information 151 from the folder 60. Further, the control unit 11 of the server 3 extracts the corrected image feature information 141 associated with the flagged image feature information 131 from the reference database of the storage unit 12 (step S109).

  Next, the control unit 11 of the server 3 compares the image histogram information of the extracted image feature information 161 with the image histogram information of the extracted image feature information 141 in the post-processing image determination unit 23, and the image feature information 141. If the similarity is recognized, a flag is set on the compared image feature information 141 (step S110). The comparison here is the same as the comparison between the image feature information 151 before the automatic correction process and the image feature information 131 described above (see FIG. 10B).

  The extracted image feature information 161 is processed in step S110 until it is compared with all the extracted image feature information 141 (step S112).

  If it is determined in step S103 that there is similarity between one piece of image feature information 151 and a plurality of pieces of image feature information 131, a plurality of pieces of image features for one piece of image feature information 161 are determined in step S109. Information 141 is extracted, and items are compared between one image feature information 161 and a plurality of image feature information 141 in step S110. In this case, all image feature information 141 is compared with items, and if there is at least one similar feature information, it may be determined that it corresponds to the optimum automatic correction process, or target image data that cannot be determined. May be determined as not optimal.

  When the process of step S110 is completed (Yes in step S111), the control unit 11 of the server 3 determines whether or not a flag exists in the image feature information 141 (step S113).

  When there is no flag in the image feature information 141 (No in step S113), there is a possibility that the corresponding target image data 51 has not obtained the optimum automatic correction processing result by the determination result output unit 24 of the server 3. The image data 81 (see FIG. 11) is displayed on the display unit 14 (step S114). Subsequently, the control unit 11 of the server 3 resets the flag of the image feature information 131 (step S117), and the i + 1th target The same processing is performed from step S103 on the image feature information 151 corresponding to the image data 51 (step S108).

  If there is a flag in the image feature information 141 (Yes in step S113), the control unit 11 of the server 3 resets the flag of the image feature information 131 and the image feature information 141 (step S117), and the next i + 1-th image. The same processing is performed from step S103 on the feature information 151 (step S108). When the process of step S113 is completed for the k-th image feature information 152 (Yes in step S115), the control unit 11 of the server 3 uses the determination result output unit 24 to perform comprehensive processing based on steps S107 and S114. The determination result is displayed on the display unit 14 (step S116).

As a comprehensive determination result, for example, the target image data 51 and the target image data 61 can be sorted and displayed as follows.
(1) The target image data 51 corresponding to the image feature information 151 before the automatic correction process that is not flagged in step S103 does not include similar image feature information 131 in the reference image data 31, that is, Sort as a new type of image with no image processing experience by the operator.
(2) The target image data 61 corresponding to the image feature information 161 after the automatic correction processing that was flagged in step S103 but not flagged in step S110 is similar to the image feature in the reference image data 41. The information 141 is classified as an image that does not exist, that is, an image that has been subjected to an inappropriate automatic correction process.
(3) The target image data 61 corresponding to the image feature information 161 flagged in both step S103 and step S110 is classified as an image that has been appropriately corrected by the automatic correction process.

Further, as shown in FIG. 11, the above classification (1) and (2) may be displayed as image data 81 in which the optimum automatic correction processing result may not be obtained. This result display requires the user of the correction quality check system 1 to check the correction quality of these image data by the operator, and depending on the correction quality, the operator may need to perform appropriate correction processing on the image data again. Is a reminder.
Further, the sorted image data (3) may be newly added to the reference database of the storage unit 12 as having been subjected to appropriate image processing, and used as a reference image for the subsequent correction quality check.
In addition, instead of displaying the results in steps S107 and S114, the results may be displayed together in step S116.

  As described above, in the present embodiment and the correction quality check system 1, it is possible to determine the correction quality of the target image that has been subjected to the automatic correction process using the image that has been corrected appropriately by the operator in the past as a reference image. As a result of the determination, the target image can be classified into an image that has been appropriately corrected by the automatic correction process and an image that requires correction quality check by the operator and appropriate correction processing. Therefore, the operator only needs to confirm what the correction quality check or appropriate correction processing is necessary, so it is not necessary to check every target image one by one, greatly improving the work efficiency of the correction quality check. To do.

  The preferred embodiments of the correction quality check system 1 and the like according to the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It will be apparent to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these are naturally within the technical scope of the present invention. Understood.

DESCRIPTION OF SYMBOLS 1 ... Correction quality check system 3 ... Server 5 ... Operation terminal 11 ... Control part 12 ... Storage part 13 ... Input part 14 ...・ Display unit 21... Image feature calculation means 22... Pre-processing image determination means 23... Post-processing image determination means 24. Reference image data 41... Reference image data 51 after correction processing... Target image data 61 before automatic correction processing... Target image data 131 after automatic correction processing. .. Multidimensional approximation function curve 134... Upper threshold value 135... Threshold lower limit 141... Image feature information 143... Multidimensional approximation function curve 144. Lower limit 151 of threshold value ... image feature information 153 ... multidimensional near Curve of the curve 161 ... image feature information 163 ... multi-dimensional approximation of the function

Claims (8)

Storage means for storing image feature information before color correction of the image data and image feature information after color correction obtained by appropriately color-correcting the image data in association with each other for a plurality of image data;
Image data input means for inputting target image data and corrected image data obtained by automatically color correcting the target image data;
Image feature calculation means for calculating image feature information for each of the input target image data and the corrected image data;
First determination means for determining whether image feature information before color correction similar to the image feature information of the target image data exists;
If it is determined that the image feature information exists, it is determined whether the image feature information after color correction corresponding to the image feature information before color correction similar to the image feature information of the corrected image data is similar. A second determination means;
An output that outputs information indicating that automatic color correction of the corrected image data is not appropriate when it is determined that the first determination means does not exist or when the second determination means determines that they are not similar Means,
A correction quality check apparatus comprising: The image feature information includes at least one piece of information indicating image data features such as histogram information, black point information, white point information, contrast information, or spatial frequency information of the image data. The correction quality check apparatus according to 1. The correction quality according to claim 1, wherein the first determination unit and the second determination unit determine a similarity by providing a threshold for the image feature information. Check device. The image characteristic information after color correction stored in the storage means is information of a result that is recognized as an appropriate color correction performed on the image data. The correction | amendment quality check apparatus in any one of. The correction quality check device
A plurality of pairs of the target image data and the corrected image data are input by the image data input means,
5. The display device according to claim 1, further comprising display means for displaying a list of the target image data or the corrected image data to which information indicating that automatic color correction is not appropriate is output. Correction quality check device. A computer comprising storage means for storing, in association with a plurality of image data, image feature information before color correction of the image data and image feature information after color correction obtained by appropriately color-correcting the image data. But,
An image data input step for inputting target image data and corrected image data obtained by automatically color correcting the target image data;
An image feature calculating step for calculating image feature information for each of the input target image data and the corrected image data;
A first determination step of determining whether image feature information before color correction similar to the image feature information of the target image data exists;
If it is determined that the image feature information exists, it is determined whether the image feature information after color correction corresponding to the image feature information before color correction similar to the image feature information of the corrected image data is similar. A second determination step;
An output that outputs information indicating that automatic color correction of the corrected image data is not appropriate when it is determined that the first determination means does not exist or when the second determination means determines that they are not similar Steps,
The correction quality check method characterized by performing. Computer
Storage means for storing image feature information before color correction of the image data and image feature information after color correction in which the image data is appropriately color-corrected in association with each other for a plurality of image data,
Image data input means for inputting target image data and corrected image data obtained by automatically color correcting the target image data;
Image feature calculation means for calculating image feature information for each of the input target image data and the corrected image data;
First determination means for determining whether or not image feature information before color correction similar to the image feature information of the target image data exists;
If it is determined that the image feature information exists, it is determined whether the image feature information after color correction corresponding to the image feature information before color correction similar to the image feature information of the corrected image data is similar. Second determination means,
An output that outputs information indicating that automatic color correction of the corrected image data is not appropriate when it is determined that the first determination means does not exist or when the second determination means determines that they are not similar means,
Correction quality check program to function as A correction quality check system in which an operation terminal and a server are connected via a network,
The operation terminal is
Input means for inputting target image data and corrected image data obtained by automatically color correcting the target image data, and transmitting the input image data to the server;
Display means for displaying the output of the server;
With
The server
Storage means for storing image feature information before color correction of the image data and image feature information after color correction obtained by appropriately color-correcting the image data in association with each other for a plurality of image data;
Means for receiving the target image data and the corrected image data;
Image feature calculating means for calculating image feature information for each of the received target image data and the corrected image data;
First determination means for determining whether image feature information before color correction similar to the image feature information of the target image data exists;
If it is determined that the image feature information exists, it is determined whether the image feature information after color correction corresponding to the image feature information before color correction similar to the image feature information of the corrected image data is similar. A second determination means;
An output that outputs information indicating that automatic color correction of the corrected image data is not appropriate when it is determined that the first determination means does not exist or when the second determination means determines that they are not similar Means,
A correction quality check system characterized by comprising:

Images