Image processing device, image processing method, and image processing program

Abstract

This invention provides an image processing device, an image processing method, and an image processing program that enable individuals with color vision deficiency to distinguish between different colors through simulation and clustering. [Solution] The image processing device is an image processing device that converts input image data input to an input device into output image data within a set color gamut on an output device, and includes: a simulation image generation unit that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion unit that converts the simulation image generated by the simulation image generation unit into an XYZ color system; a confusion color information storage unit that stores confusion color information relating to confusion color lines that show the distribution of colors that are difficult for colorblind persons to distinguish; and a clustering processing unit that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors.

Description

[Technical Field]

[0001] The present invention relates to an image processing apparatus, an image processing method, and an image processing program, and more particularly to an image processing apparatus, an image processing method, and an image processing program capable of providing images in which even people with color vision deficiency can distinguish color differences. [Background technology]

[0002] People with normal vision can see colors such as red. However, people with color blindness, color vision deficiency, or other color vision impairments cannot accurately perceive colors such as red. Humans can see things when light illuminates an object and the reflected light is received by the eye. Different objects reflect different wavelengths of light, which creates the difference in their color. The retina of the human eye contains two types of photoreceptor cells: rods, which function only in the dark, and cones, which function only in bright light. There are three types of cones: red, blue, and green, each playing a role in sensing light of different wavelengths. The stimuli received by these cones are then transmitted to the cerebrum, where color is recognized.

[0003] Cone cells are determined at birth, with the most common being type C, which has three types of cones: red, green, and blue. This type accounts for 95% of Japanese men and over 99% of Japanese women. The next most common is type D, which either lacks green cones or perceives wavelengths similar to red. Following that is type P, which either lacks red cones or perceives wavelengths similar to green. Type C individuals are called "normal color vision" people, while types D and P individuals are called "color blind" people.

[0004] Therefore, an image processing device has been proposed that can output image data that sufficiently maintains the ease of identification of color schemes for people with color vision deficiency (see, for example, Patent Document 1). The image processing apparatus disclosed in Patent Document 1 includes an input color information conversion unit that converts image data in an input device into an XYZ color space system, a confusion color information storage unit that stores confusion color information regarding a confusion color line indicating a distribution of colors that are difficult for color vision-deficient individuals to distinguish, a color gamut mapping unit that uses the confusion color information to convert the color of the image data in the XYZ color space system obtained by the input color information conversion unit into XYZ values within the color gamut of an output device so that the color does not lie on the same confusion color line, and an output color information conversion unit that converts the image data obtained by the color gamut mapping unit into the color space system of the output device.

[0005] However, in the image processing apparatus disclosed in Patent Document 1, color-deficient individuals did not accurately grasp how colors actually appeared. Also, clustering that classifies a plurality of similar colors and aggregates them into a representative number of colors was not performed. Therefore, there has been a desire for the emergence of a technology that enables color-deficient individuals to accurately distinguish color differences without confusing different colors as the same color by using clusters obtained by classifying a plurality of similar colors by clustering based on the colors that color-deficient individuals actually see and aggregating them into a representative number of colors.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0007] Therefore, the present invention has been made in view of the above points, and provides an image processing apparatus, an image processing method, and an image processing program capable of providing an image in which color differences can be distinguished even by color vision-deficient individuals through simulation and clustering.

Means for Solving the Problems

[0008] In other words, the image processing apparatus according to the first embodiment is an image processing apparatus that converts input image data input to the input device side into output image data within the color gamut set on the output device side, The system is characterized by comprising: a simulation image generation unit that generates a color vision simulation image by performing a dichromacy simulation on input image data in an input device; an input color information conversion unit that converts the simulation image generated by the simulation image generation unit into an XYZ color system; a color confusion information storage unit that stores color confusion information relating to color confusion lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish; a clustering processing unit that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction unit that uses the color confusion information in the color confusion information storage unit to perform color correction by moving the colors of the XYZ color system image data obtained by the input color information conversion unit to a second color confusion line that is separated from a first color confusion line to which one of the clusters obtained by clustering is located, on a straight line connecting multiple clusters obtained by clustering, without matching it with the color confusion line which is a straight line formed from points where multiple mixed colors in the XYZ color system are plotted; and an output color information conversion unit that converts the image data obtained by the color correction unit into the color system of an output device.

[0009] The image processing apparatus according to the second embodiment is an image processing apparatus that converts input image data input to an input device into output image data within a color gamut set on an output device, and may include: a simulation image generation unit that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion unit that converts the simulation image generated by the simulation image generation unit into a Luv color system; a clustering processing unit that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction unit that performs color correction by moving the colors of the Luv color system image data obtained by the input color information conversion unit onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to a cluster on a straight line connecting multiple clusters obtained by clustering; and an output color information conversion unit that converts the image data obtained by the color correction unit into a color system of an output device.

[0010] A third embodiment is an image processing apparatus according to the first or second embodiment, wherein the image data in the input device is color-coded to be easily distinguishable by a person with color vision deficiency.

[0011] A fourth embodiment is an image processing apparatus according to the first or second embodiment in which dichromacy simulation may be performed by reducing RGB, converting RGB to LMS, interpolating the L deficiency with MS, and converting LMS back to RGB.

[0012] A fifth embodiment may be an image processing apparatus according to the first or second embodiment, comprising: a rotational image transfer unit that performs a rotational mapping on the XYZ color system image data obtained by the input color information conversion unit, with respect to the concentration points of the confused color lines; and a parallel mapping unit that performs a horizontal mapping on the confused color lines.

[0013] The sixth embodiment is an image processing apparatus according to the first or second embodiment, in which the amount of image in the re-transferred image may be determined by the ratio of the color gamut of the input device to the color gamut of the output device.

[0014] The seventh aspect of the image processing method is an image processing method that converts input image data input to an input device into output image data within a set color gamut on an output device, comprising: a simulation image generation step of generating a color vision simulation image by performing a dichromacy simulation on the input image data in the input device; and an input color information conversion step of converting the simulation image generated in the simulation image generation step into an XYZ color system. The method is characterized by comprising: a color confusion information storage step for storing color confusion information relating to color confusion lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish; a clustering processing step for performing clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction step for performing color correction using the color confusion information from the color confusion information storage step, by moving one of the clusters obtained by clustering onto a second color confusion line that is spaced apart from a first color confusion line to which one of the clusters coincides or is close, on a straight line connecting multiple clusters obtained by clustering, without making it coincide with the mixed color line which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system obtained in the input color information conversion step; and an output color information conversion step for converting the image data obtained in the color correction step into the color system of the output device.

[0015] The eighth aspect of the image processing method is an image processing method that converts input image data input to an input device into output image data within a color gamut set on an output device, comprising: a simulation image generation step of generating a color vision simulation image by performing a dichromacy simulation on the input image data in the input device; an input color information conversion step of converting the simulation image generated in the simulation image generation step into a Luv color system; a clustering processing step of performing clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction step of performing color correction by moving the colors of the Luv color system image data obtained in the input color information conversion step onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to a cluster, on a straight line connecting multiple clusters obtained by clustering; and an output color information conversion step of converting the image data obtained in the color correction step into the color system of the output device.

[0016] The image processing program according to the ninth aspect is a program that causes a computer to execute an image processing function that converts input image data input to an input device into output image data within a set color gamut on an output device, comprising: a simulation image generation function that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion function that converts the simulation image generated by the simulation image generation function into an XYZ color system; a confusion color information storage function that stores confusion color information relating to confusion color lines that show the distribution of colors that are difficult for colorblind persons to distinguish; and a plurality of similar The system is characterized by performing the following functions: a clustering processing function that classifies colors and aggregates them into a representative number of colors; a color correction function that uses the mixed color information stored by the mixed color information storage function to correct colors by moving a cluster to a second mixed color line that is separated from a first mixed color line to which one of the clusters obtained by clustering is located, on a straight line connecting multiple clusters obtained by clustering, without making it coincide with the mixed color line which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system obtained by the input color information conversion function; and an output color information conversion function that converts the image data obtained by the color correction function into the color system of the output device.

[0017] The image processing program according to the tenth embodiment is a program that causes a computer to execute an image processing function that converts input image data input to an input device into output image data within a set color gamut on an output device, characterized in that it executes: a simulation image generation function that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion function that converts the simulation image generated by the simulation image generation function into a Luv color system; a clustering processing function that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction function that performs color correction by moving the colors of the Luv color system image data obtained by the input color information conversion function onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to a cluster on a straight line connecting multiple clusters obtained by clustering; and an output color information conversion function that converts the image data obtained by the color correction function into the color system of the output device. [Effects of the Invention]

[0018] The image processing apparatus according to the present invention is an image processing apparatus that converts input image data input to an input device into output image data within a color gamut set on an output device, comprising: a simulation image generation unit that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion unit that converts the simulation image generated by the simulation image generation unit into an XYZ color system; a confusion color information storage unit that stores confusion color information relating to confusion color lines that show the distribution of colors that are difficult for color-blind persons to distinguish; and a clustering processing unit that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors. The system is characterized by having a color correction unit that uses the mixed color information of the mixed color information storage unit to move the color of the XYZ color system image data obtained by the input color information conversion unit to a second mixed color line that is spaced apart from a first mixed color line to which one of the clusters obtained by clustering processing is located, on a straight line connecting multiple clusters obtained by clustering processing, without matching it with the mixed color line which is a straight line formed from points where multiple mixed colors are plotted in the XYZ color system; and an output color information conversion unit that converts the image data obtained by the color correction unit into the color system of the output device. As a result, even people with color vision deficiency can distinguish between colors through simulation and clustering.

[0019] Furthermore, the image processing method and image processing program according to the present invention, like the image processing apparatus according to the present invention, enable color-blind individuals to distinguish color differences through simulation and clustering. [Brief explanation of the drawing]

[0020] [Figure 1] This is a block diagram showing the schematic configuration of an image processing apparatus according to an embodiment. [Figure 2] This is a diagram showing the functional configuration of an image processing apparatus according to an embodiment. [Figure 3] This is a block diagram showing the functional configuration of the color correction unit of the image processing apparatus according to the embodiment. [Figure 4](a) This figure shows how colors are perceived by people with normal color vision. (b) This figure shows how colors are perceived by people with protanopia (P-type color vision deficiency). (c) This figure shows how colors are perceived by people with dyschromia (D-type color vision deficiency). [Figure 5] This is a flowchart illustrating the process of automatic color adjustment in the embodiment. [Figure 6] This is a flowchart of the K-means++ algorithm for automatic color scheme adjustment in the embodiment. [Figure 7] This is a flowchart of the gradient descent algorithm for automatic color scheme adjustment in the embodiment. [Figure 8] This figure shows a simulation of type P-color vision deficiency according to the embodiment. [Figure 9] This is a graph showing the LMS cone response of the embodiment. [Figure 10] This figure shows the clustering process of the embodiment. [Figure 11] This figure shows an example of automatic color adjustment in the XYZ color system of the embodiment. [Figure 12] This figure shows an example of color correction in the XYZ color system of the embodiment. [Figure 13] (a) This is a trichromatic vision diagram. (b) This is a dichromatic vision diagram (P-type). [Figure 14] This figure shows an example of color correction in Luv space according to the embodiment. [Figure 15] (a) This is a trichromatic vision diagram. (b) This is a dichromatic vision diagram (P-type). [Figure 16] This figure shows a sample image with automatic color adjustment according to the embodiment. [Figure 17] This figure shows the screen configuration for automatic color adjustment in the embodiment. [Figure 18] This figure shows the preview screen of the drawing application of the embodiment. [Figure 19] This is an example of a flowchart for an image processing method according to an embodiment. [Figure 20] This is an example of a flowchart for an image processing method of another embodiment. [Modes for carrying out the invention]

[0021] An example of the configuration of the image processing apparatus 1 according to the embodiment will be described with reference to Figures 1 and 2. Figure 1 is a block diagram showing an example of the schematic configuration of the image processing apparatus 1, and Figure 2 is a block diagram showing an example of the functional configuration of the image processing apparatus 1.

[0022] As shown in Figure 1, the image processing device 1 has a CPU 6 (a computing element such as a GPU), and the CPU 6 is connected to an input / output interface 5, a program ROM 7, RAM 8, and an HDD 9 (SSD). The program ROM 7 stores a program related to image processing, and the CPU 6 loads the image processing program stored in the program ROM 7 onto RAM 8 and executes the program, thereby performing the processing operations of the image processing device 1 described later. The HDD 9 is used as a storage medium to store various data necessary for controlling the operation of the image processing device 1, such as images created by the user for people with color blindness, color gamut information of the monitor 3, color gamut information of the printer 4, and the color vision characteristics of people with color blindness. In addition, the input / output interface 5 is connected to the monitor 3 and keyboard 2 as input devices, and to the printer 4 as an output device. In the input devices, the image data is color-coded to be easily distinguishable by people with color blindness.

[0023] Figure 2 is a functional block diagram of the image processing apparatus 1 of this embodiment. The image processing apparatus 1 of this embodiment includes a simulation image generation unit 10, an input color information conversion unit 20, a confused color information storage unit 30, a clustering processing unit 40, a color correction unit 50, and an output color information conversion unit 60. These are functional units that are realized by the CPU 6 executing an image processing program deployed on RAM 8.

[0024] The simulation image generation unit 10 generates a color vision simulation image by performing a dichromacy simulation on the input image data from the input device. The dichromacy simulation will be described in detail later.

[0025] The input color information conversion unit 20 converts the simulation image generated by the simulation image generation unit 10 into either the XYZ color system or the Luv color system. The XYZ color system and the Luv color system will be described in detail later.

[0026] The color confusion information storage unit 30 stores color confusion information relating to color confusion lines, which show the distribution of colors that are difficult for people with color vision deficiency to distinguish. A color confusion line is a line segment on which points that lie on the same line all appear to be the same color, as shown in Figure 9.

[0027] The clustering processing unit 40 performs clustering, classifying multiple similar colors and consolidating them into a representative number of colors. Clustering will be explained in detail later.

[0028] The color correction unit 50 uses the mixed color information from the mixed color information storage unit 30 to perform color correction by moving the colors of the XYZ color system image data obtained by the input color information conversion unit 20 to a second mixed color line that is spaced apart from a first mixed color line where one cluster coincides or is close to, on a straight line connecting multiple clusters obtained by clustering, instead of matching it with the mixed color line, which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system. The color correction unit 50 also performs color correction by moving the colors of the Luv color system image data obtained by the input color information conversion function to a second mixed color line that is spaced apart from a first mixed color line where one cluster coincides or is close to, on a straight line connecting multiple clusters obtained by clustering.

[0029] Furthermore, as shown in Figure 3, the color correction unit 50 includes a rotational image transfer unit 51 and a parallel mapping unit 52. The rotational image transfer unit 51 maps the XYZ color system image data obtained by the input color information conversion unit in the rotational direction, centered on the confusion line concentration point where the confusion lines are concentrated. The parallel mapping unit 52 maps the confusion lines in the horizontal direction. The amount of mapping in the rotational image transfer is determined by the ratio of the color gamut of the input device to the color gamut of the output device.

[0030] The output color information conversion unit 60 converts the image data obtained by the color correction unit 50 into the color system of the output device.

[0031] Figure 4 shows an example of a color vision diversity simulation image. For illustrative purposes, the image is represented in black and white grayscale. In reality, the image is colorized. The same applies to subsequent grayscale drawings. (a) is the original image showing how it looks to a person with normal color vision, i.e., a person with normal color vision. The number "8" is shown in red against a background of green and black. (b) is a diagram showing how it looks to a person with protanopia (type 1). In (b), the number "8" is embedded in the background of green and black. (c) is a diagram showing how it looks to a person with deuteranopia (type 2). It is easier to see than in (b), but the number "8" is almost completely embedded in the background of green and black.

[0032] A simulator can reveal how people with dichromacy (color vision deficiency) perceive things. Figure 4 shows an example of an actual simulation image, and from the simulation image, it can be inferred that people with protanopia (P-type) and deuteranopia (D-type) color vision deficiency have difficulty recognizing the letter "8". Automatic color adjustment is a function that automatically adjusts the original image to a color scheme that can be distinguished by people with P-type / D-type color vision deficiency.

[0033] People with color vision deficiency see colors differently than those with all three types of cones (photoreceptor cells) in the retina of the human eye because they lack some or all of them, or have a misalignment of their spectral sensitivity. Common type (Type C) is diagnosed as having normal color vision by an ophthalmologist and possesses all three types of cones; it accounts for approximately 95% of Japanese men and over 99% of Japanese women. Protanopia type (Type P) is characterized by the absence or misalignment of cones that primarily sense red light. Deuteranopia type (Type D) is characterized by the absence or misalignment of cones that primarily sense green light. Types P and D have similar visual impairments, making it difficult to distinguish between colors near red and colors near green. Most people with color vision deficiency fall into one of these two types.

[0034] FIG. 5 is a flowchart showing the automatic color matching adjustment process of the image processing apparatus 1 according to the embodiment. The original image is converted from both the original image and the image subjected to dichromatic vision simulation into the CIE L in the uniform color space * u * v * space by clustering. Based on the classified cluster center coordinates, the clusters to be determined for the confusion colors to be judged and corrected are determined. The correction cluster is color-corrected by confusion color correction to complete the color matching adjustment. The confusion color correction is implemented in two types in the CIE Yxy space and the CIE L * u * v * space.

[0035] CIE (Commission internationale de l’eclairage) is the International Commission on Illumination, an international standardization organization that defines light, illumination, color, and color spaces. In television, it is defined based on the three-dimensional visible color space CIE 1931 established in 1931. Although CIE 1931 is a three-dimensional XYZ space, the color spaces of ITU-R BT.709 (2K high definition) and ITU-R BT.2020 (4K super high definition) defined in television are defined in the CIE Yxy chromaticity diagram projected onto a two-dimensional space based on the Y axis. The CIE L * u * v * color space is a color space formulated based on Mac Adam's color value color space and represents a color space close to human perception. Clustering is a process of classifying a plurality of similar colors and aggregating them into a representative number of colors, thereby improving the calculation efficiency.

[0036] Here, the flowchart shown in FIG. 5 will be described in detail. The following description is based on Col Res Appl, Volume 33, Issue 3, pages 203-211, 2008. First, an RGB image is input. The input image is passed through the path of converting the original data into CIE L * u * v * and through dichromatic vision simulation into CIE L * u* v * The conversion is separated into two paths. This is because dichromacy simulation data is used for color confusion detection. CIE L * u * v * The converted data is clustered using the k-means++ algorithm, both for the original data and the simulation data, in order to efficiently handle multiple colors. The number of classifications is a hyperparameter. The cluster center coordinates (L) of the classified clusters are then determined. * ,u * ,v * For all combinations, a color confusion test is performed to determine color confusion clusters. All colors within the determined clusters are color-corrected using color confusion correction. Clusters that are not determined to be color confusion clusters, and the colors belonging to those clusters, are not corrected. Color confusion correction is performed by adjusting the center coordinates of the clusters that were targeted for correction after the dichromacy simulation.

[0037] Next, we will explain K-means++ using Figure 6. Here too, we will explain based on the previously mentioned paper. Figure 6 is a flowchart of the K-means++ algorithm. In step S1, the data point p(L * ,u * ,v * Let any point on ) be the center C1. Next, in step S2, the data point p(L * ,u * ,v * ) A new center Ci is determined from the distance to the center already selected. If Yes in step S3, proceed to step S4. If No in step S3, return to step S2. In step S4, calculate the distance between all points and the K centers and update the classification label to the closest center. In step S5, calculate the center from each of the K classified data and set that center as Cinew. If Yes in step S6, the process ends, and if No in step S6, return to step S4.

[0038] Next, we will explain color confusion detection. Color confusion detection is performed using cluster center coordinates Ci(L * ,u * ,v* The aforementioned method from the paper is used to calculate the color difference ΔEuv between the two groups from the ratio of the trichromacy difference ΔEn and the dichromacy difference ΔEd. For example, the color difference between the trichromacy clusters C1n and C2n and the dichromacy clusters C1d and C2d is calculated as follows. First, ΔEn and ΔEd are calculated from the following formulas.

[0039]

number

[0040] If ΔEn and ΔEd simultaneously satisfy the following conditions 1 and 2 shown in the equation below, the two clusters are determined to be confused colors.

[0041]

number

[0042] Based on the method described in the aforementioned paper, the values ​​r=0.9 and ΔEdmax=50 are used, and the software is implemented using these values. If clusters 1 and 2 are found to be confused colors in the color confusion detection, color correction will be applied to both clusters. Correcting either one of the clusters will remove it from the confused color category, but the cluster with fewer colors will be selected as the corrected cluster. This is to reduce the amount of color correction computation.

[0043] Next, we will explain color confusion correction. Color confusion correction is implemented by optimizing the cluster to be corrected using an iterative gradient descent method. Gradient descent requires defining an evaluation function. The degree of confusion is defined as the reciprocal of the L2 norm between the cluster to be corrected Ci and the other clusters Ck, and the smaller this value, the less confused the clusters are. Therefore, the degree of confusion Φconf(Ci) is defined by the following formula.

[0044]

number

[0045] Here, Ck(k=1,2,·····,N) represents the cluster centers, and Ci(i is a subset of {1,2,······,N}) represents the modified clusters. Both represent the simulated clusters. Also, ε is a small number to prevent division by zero. Next, the degree of cluster modification Φsim(Ci) is defined as follows:

[0046]

number

[0047] Here, Cio is the original cluster center before modification. Minimizing the amount of modification to the original contributes to a good modification. To ensure a confused color cluster center, we must minimize the sum of the two terms defined by equations 7 and 8.

[0048]

number

[0049] Here, λ is a positive number used to adjust the effective balance of the two terms, and we use λ = 0.0005. Therefore, the evaluation function E is as follows:

[0050]

number

[0051] To find the optimal value of Ci from this evaluation function using gradient descent, it is necessary to calculate the change ΔCi, which is obtained by differentiating E with respect to Ci. This derivative ΔCi is given by the following equation.

[0052]

number

[0053] The optimal value is obtained by an iterative correction method in which a small amount of ΔCi (Ci ← Ci + ΔCi) is added so that each cluster center Ci reduces the evaluation function. Although the paper describes the assumption of correcting multiple cluster centers, only one cluster is corrected in the current software implementation.

[0054] Next, the RGB correction image output will be described. CIE L * u * v * The data automatically color-adjusted in the space is converted to the RGB space, processed by the OETF (Optical Electrical Transfer Function) of ITU-R BT.709, and output as 8-bit RGB image data.

[0055] Next, referring to Figure 7, the gradient descent method will be described. Figure 7 is the algorithm flowchart of the gradient descent method. Color correction optimization is a method of finding the minimum value of the evaluation function by iterative correction using the general gradient descent method. An initial value is set for x in step S7, and Δx = f'(x) in step S8. Next, Δx < th in step S9. If Yes in step S9, the process ends. If No in step S9, it proceeds to step S11. Step S11 is xn+1 ← xn - ηΔx, and it returns to step S8.

[0056] Assuming the evaluation function is f(x), the gradient descent method for minimum value detection is as shown in Figure 8. Here, η is the learning rate and contributes to the accuracy of the optimal value x. Also, th is the threshold value, which is the condition for exiting the iterative processing loop. The value of x when Δx < th holds brings the minimum value of the evaluation function f(x). Looking at the following evaluation differential value formula.

[0057]

Equation

[0058] As shown above, ΔCi is not proportional to ∂E / ∂Ci, but rather proportional. However, the correction is set to Ci ← Ci + ΔCi, and the iteration process is terminated when ΔCi < 0.01 occurs three times in a row. Although I actually performed the calculation, I was unable to reproduce the color correction that resolves the confused colors in the Ishihara Plate 02 image in 5 iterations, as described in the paper.

[0059] Refer to Figures 8 and 9 to explain the dichromacy simulation. The dichromacy simulation implements the P-type simulation model introduced in Francoise Vienot's paper "Digital Video Colourmaps for Checking the Legibility of Display by Dichromats". The P-type simulation is realized by the calculations shown in Figure 5. The RGB data undergoes degenerate processing to prevent overflow during processing, and the LMS conversion is a color space that mimics human cone responses, consisting of Long (long wavelength), Medium (medium wavelength), and Short (short wavelength), where L corresponds to red, M to green, and S to blue. The LMS cone responses are shown in Figure 9. The P-type simulation is performed using L-deletion interpolation, which interpolates the signal lost due to the L defect using the remaining two cone responses, M and S. The P-type simulation is completed by converting the LMS space, which has undergone simulation processing, back to the RGB space.

[0060] Next, we will explain the diagram shown in Figure 8 in detail. Linear RGB data undergoes the following reduction process to suppress the occurrence of Gamato errors during processing in Step 1.

[0061]

number

[0062] Next, the Judd-Vos colorimetric correction and the Smith and Pokorny matrix are used to convert RGB to LMS using the following formula.

[0063]

number

[0064] Next, we perform an L-deficient dichromacy simulation using the following matrix derived from the LM plane equations for primary color and reference white in ITU-R BT.709.

[0065]

number

[0066] Next, convert LMS to RGB using the following formula.

[0067]

number

[0068] The obtained [Rp Gp Bp]T is P-type simulated RGB data. The RGB data is processed through CIE XYZ space, and then CIE L * u * v * Convert into space.

[0069] Figure 10 shows a clustering diagram. Clustering is CIE L * u * v * By classifying many colors in space, color confusion and color correction processing can be performed efficiently. Figure 8 shows an image of three classified clustering processes. Here, the number of classifications, 3, is a hyperparameter. A hyperparameter is a value determined by adjustment, and in the processing of this embodiment, the number of classifications cannot be determined from the data, so it is determined while adjusting. In the implemented software, the number of classifications is fixed at 4. In the initial state shown in the left figure (bottom of the page), countless white "x"s are distributed around the part that looks like a white mass. * u * v * Each color in space is represented by three circles scattered on the white, block-like areas, which indicate the initial values ​​of the cluster centers. In the clustering process, L * u* v * Each color is classified into three categories through an iterative process so that the straight-line distance between each color in space is average. As a result, the white "x" symbols, which appear as white clumps, are distributed in countless numbers around the area and are classified into clusters colored red, blue, and green, each representing a different classification, and the "o" symbols move to the central coordinate position of each cluster. Figure 10 shows the classification display in 2D (u * ,v * Although it is shown as ) in reality it is (L * ,u * ,v * The classification process is performed in 3D.

[0070] Figure 11 shows an example of automatic color adjustment in the XYZ color system. A color correction method based on the confusion lines in the CIE Yxy space described in the Color Universal Design Guide is implemented. This method performs color matching using a geometric method in the CIE Yxy space. This is the conversion method for the Ishihara Plate 02 image. Cin1~3 represent clusters in the original image, and Cid1~3 represent clusters in the Vienot P-type simulation image. f475RefWhite is the straight line projected onto a P-type dichromacy. Cin3 is close to the confusion line ftypP(Cin2) of Cin2 and is judged to be a confused color, so Cin3 is used as the correction cluster. To correct the cluster of Cin3, first the simulation result Cid3 is placed on the opposite side of the reference white on f475RefWhite (Cid2New). A confusion line ftypP(Cid2New) is drawn connecting Cid2New and the P-type confusion center (x:y)(0.26531:0.73469). Furthermore, a line fn475(Cin2) is drawn through Cin2 and parallel to f475RefWhite. The intersection of the two lines ftypP(Cid2New) and fn475(Cin2) becomes the new cluster position Cin2New, which is modified from Cin2.

[0071] Refer to Figures 12 and 13 to see examples of color correction in the XYZ color system. In color correction using the CIE Yxy space, as shown in Figure 9, all clusters of trichromats on the confusion line are aggregated onto the P-type simulation line by the P-type simulation (see the dashed line in the figure). For example, if there are clusters C1d (thin solid line in the figure), C2d (thick dashed line in the figure), and C3d (thick dashed line in the figure) aggregated by the simulation, and C2d is a confused color (the color difference distance between C1d and C2d is small), color correction is performed by moving C2d to a location far from C1d (C2d_new). Also, Figure 10 shows the trichromatic and P-type dichromatic colors in the CIE Yxy chromaticity diagram. (a) is the trichromatic chromaticity diagram, and (b) is the P-type dichromatic chromaticity diagram. For P-type colorblind individuals, it is not possible to see in the same way as a trichromat, but they can recognize that the upper left part and the lower right part are different colors.

[0072] Refer to Figures 14 and 15 to see examples of color correction in the Luv color system. CIE L * u * v * Spatial color correction is performed on clusters that converge on a P-type simulation line, as shown in Figure 14. As shown in Figure 14, there are clusters C1d, C2d, and C3d. If C2d is a confused color (the color difference distance between C1d and C2d is small), color correction is performed by moving C2d to a location far from C1d (C2d_new). Also, CIE L * u * v * Figure 15 shows the color chromaticity diagrams for trichromats and protanopia. (a) is the trichromaticity diagram, and (b) is the protanopia color chromaticity diagram. It is difficult to see because it cannot be represented in color, but for protanopia, although they cannot see it the same way as trichromats, they can recognize that the upper left and lower right parts are different colors.

[0073] Figure 16 shows sample images of the automatic color adjustment embodiment. The transition from the first to the second row is the result of automatic color adjustment using the XYZ color system described above. (a) is the original image, and since it is how a person with normal color vision would see it, the number "8" is clearly distinguishable from the background color. (b) is the original image with a P-type simulation applied. The number "8" is almost indistinguishable from the background color. (c) is the original image with a D-type simulation applied. Again, the number "8" is almost indistinguishable from the background color. (d) is the image with color adjustment using the XYZ color system. Since it is how a person with normal color vision would see it, the number "8" is clearly distinguishable from the background color. (e) is the image with a P-type simulation applied using the XYZ color system. Since color adjustment has been performed, it is not the same color as (d), but compared to (b), the number "8" is slightly distinguishable from the background color. (f) is the image with a D-type simulation applied using the XYZ color system. After adjusting the color scheme, the number "8" is now quite clearly distinguishable from the background color, although it's not the same color as (d).

[0074] The transition from the first to the third row shows examples of images with the aforementioned automatic color adjustment in Luv space. (g) is an image with color adjustment in Luv space. As this is how it looks to a person with normal color vision, the number "8" is clearly distinguishable from the background color. (h) is a P-type simulation image with color adjustment in Luv space. Because color adjustment has been performed, it is not the same color as (g), but compared to (b), the number "8" is slightly distinguishable from the background color. (i) is a D-type simulation image with color adjustment in Luv space. Because color adjustment has been performed, it is not the same color as (g), but the number "8" is much more distinguishable from the background color.

[0075] Figure 17 shows the screen configuration for automatic color adjustment in the embodiment. The upper section shows the state before color adjustment, and the lower section shows the state after color adjustment. The left side of the upper section shows how it looks to people with normal color vision and trichromacy. The center shows how it looks to people with dichromacy and protanopia. The right side shows how it looks to people with dichromacy and dechromacy. Similarly, the lower section shows how it looks to people with normal color vision and trichromacy on the left side, how it looks to people with dichromacy and protanopia on the center, and how it looks to people with dichromacy and dechromacy on the right side. It can be seen that in the lower section, the colors on the left side of the graph are clearer and can be distinguished from other colors.

[0076] This color scheme adjustment preview screen offers various pane view functions. First, there's the zoom in / zoom out function. It allows you to focus on a specific area and provides a zoom function to view the overall picture. The synchronized pane view function makes it easy to check and compare the same position. Next, there's the grab scroll function. It provides a function that allows you to easily move the display area by grabbing the pane view. The synchronized pane view display function makes it easy to check and compare the same position. Next, there's an interactive color confirmation tool. It provides a color picker function to check the colors (RGB values) before and after color scheme adjustment. Next, there's a color vision simulation. It provides a function to simulate how images appear to people with color vision deficiencies (protanopia, deltoid).

[0077] The toolbar includes icons that perform the following functions: an icon to display color adjustment information, showing the original and adjusted colors in RGB values; an icon to output the color-adjusted image and specify the output folder; an icon to toggle the display of the color vision type name (blue background for before adjustment, pink background for after adjustment); an icon to toggle the display of Type C grayscale (grayscale display is only possible for Type C); an icon to toggle the display of the color-adjusted area; an icon to toggle the display of Type C grayscale; and a switch to Type C (normal color vision) display mode. Icons for switching between D-type and P-type displays (shortcuts: left / right) and P-type displays (right). Icons for switching between P-type and D-type displays (shortcuts: left / right) and D-type displays (right). Icons for switching between P-type and D-type displays (shortcuts: left / right) and D-type displays (right). Icons for increasing the zoom level by 5% and decreasing the zoom level by 5%. Icon for displaying at a size that fits within the display pane and icon for displaying at 100% pixel size (100% HD (1920x1080) display). Pressing the "Apply" button will reflect the color adjustment results in the material.

[0078] Figure 18 shows the preview screen of the drawing application of the embodiment. This window provides drawing support functions for drawing artists with knowledge of color universal design to adjust color schemes. The left side shows the hues as seen by people with normal color vision. Single-color editing of this unit can be performed, and the color being adjusted can be checked on the preview screen. The right side is the preview screen, with the bottom showing the hues as seen by people with normal color vision, the upper left showing the hues as seen by people with protanopia (P-type dichromacy), and the upper right showing the hues as seen by people with deuteranopia (D-type dichromacy). The drawing application provides the following functions on the preview screen: a function to display the color vision simulation results in real time without affecting the drawing work; a function to display the change results in the preview window in real time when the text overlay material is changed, allowing visibility to be checked; a function to check text overlay materials with motion effects as a video; a function to display the adjustment results in the preview window without applying the adjustments when adjusting the color in the single-color editing window of the text overlay unit, allowing color adjustment using RGB values, brightness (L), saturation (S), etc., while checking visibility, and also displaying the adjusted color in the preview on the text overlay unit in the editing view.

[0079] Referring to Figure 19, the image processing program according to the embodiment will be described along with the image processing method. Figure 17 is an example of a flowchart of the image processing method of the embodiment. The image processing method is executed by the CPU 6 of the image processing device 1 based on the image processing program. The image processing program includes a simulation image generation step S10, an input color information conversion step S20, a confused color information storage step S30, a clustering processing step S40, a color correction step S50, and an output color information conversion step S60, among others. The image processing program implements functions such as simulation image generation, input color information conversion, confused color information storage, clustering processing, color correction, and output color information conversion on the CPU 6 of the image processing device 1. These functions are executed in the order shown in the flowchart of Figure 18, but the order can be changed as appropriate. Since each function overlaps with the descriptions of the various functions of the image processing device 1 mentioned above, detailed explanations are omitted.

[0080] The simulation image generation function generates a color vision simulation image by performing a dichromacy simulation on the input image data in the input device (Step S10: Simulation image generation step).

[0081] The input color information conversion function converts the simulation image generated by the simulation image generation function into an XYZ color system (Step S20: Input color information conversion step).

[0082] The color confusion information storage function stores color confusion information related to color confusion lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish (Step S30: Color confusion information storage step).

[0083] The clustering function performs clustering, which classifies multiple similar colors and aggregates them into a representative number of colors (Step S40: Clustering Processing Step).

[0084] The color correction function uses the mixed color information stored by the mixed color information storage function to correct the color by moving the clusters obtained by clustering onto a second mixed color line that is separated from the first mixed color line where one cluster coincides with or is close to the first mixed color line, without making it coincide with the mixed color line, which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system obtained by the input color information conversion function (Step S50: Color correction step).

[0085] The output color information conversion function converts the image data obtained by the color correction function into the color system of the output device (Step S60: Output color information conversion step).

[0086] Next, with reference to Figure 20, an image processing program according to another embodiment will be described along with an image processing method according to another embodiment. Figure 20 is an example of a flowchart of an image processing method according to another embodiment. The flowchart of the image processing program according to another embodiment shown in Figure 20 differs from the flowchart of the image processing program shown in Figure 19 in that the step for storing confused color information is omitted. The image processing method according to the other embodiment is executed by the CPU 6 of the image processing device 1 based on the image processing program according to the other embodiment shown in Figure 20. The image processing program according to another embodiment shown in Figure 20 includes a simulation image generation step S10, an input color information conversion step S20, a clustering processing step S30, a color correction step S40, and an output color information conversion step S50, among others. The image processing program implements functions such as simulation image generation, input color information conversion, clustering processing, color correction, and output color information conversion on the CPU 6 of the image processing device 1. These functions are executed in the order shown in the flowchart of Figure 20, but the order can be changed as appropriate. Since each function overlaps with the descriptions of the various functions of the image processing device 1 mentioned above, detailed explanations are omitted.

[0087] The simulation image generation function generates a color vision simulation image by performing a dichromacy simulation on the input image data in the input device (Step S10: Simulation image generation step).

[0088] The input color information conversion function converts the simulation image generated by the simulation image generation function into the Luv color system (Step S20: Input color information conversion step).

[0089] The clustering processing function performs clustering, classifying multiple similar colors and consolidating them into a representative number of colors (Step S30: Clustering Processing Step).

[0090] The color correction function corrects the color of the Luv color system image data obtained by the input color information conversion function by moving it to a second confusion color line that is separated from a first confusion color line where one cluster coincides with or is close to a cluster, on a straight line connecting multiple clusters obtained by clustering (Step S40: Color correction step).

[0091] The output color information conversion function converts the image data obtained by the color correction function into the color system of the output device (Step S50: Output color information conversion step).

[0092] According to each aspect of this disclosure described above, even people with color blindness can distinguish color differences through simulation and clustering.

[0093] The image processing program of this embodiment can be implemented on the CPU 6 using a computer language such as the Arduino® IDE.

[0094] [Regarding functions and circuitry] Next, the functions and circuitry of the image processing device 1 described above will be explained. Each functional unit of the CPU 6 of the image processing device 1 may be implemented as a function of a computer's arithmetic processing unit or the like. That is, the CPU 6 may be implemented as a simulation image generation function, an input color information conversion function, a confused color information storage function, a clustering processing function, a color correction function, and an output color information conversion function, respectively, by a computer's arithmetic processing unit or the like. The image processing program can enable a computer to implement the functions described above. The image processing program may be recorded on a computer-readable, non-temporary storage medium, such as memory, a solid-state drive, a hard disk drive, or an optical disc. The storage medium can be rephrased as a non-temporary, computer-readable medium for storing the image processing program. Furthermore, the image processing program may be transmitted online. Furthermore, the computer's arithmetic processing unit and the like described above may be composed of, for example, integrated circuits. That is, the CPU 6 of the image processing device 1 may be implemented as a simulation image generation circuit, an input color information conversion circuit, a confused color information storage circuit, a clustering processing circuit, a color correction circuit, and an output color information conversion circuit that constitute the computer's arithmetic processing unit and the like.

[0095] Furthermore, the present invention is not limited to the image processing apparatus 1, image processing method, and image processing program according to the above-described embodiment, and can be implemented by various other modifications or applications without departing from the gist of the present invention as described in the claims. Also, although the word "information" is used in the above-described embodiment, the word "information" can be replaced with "data," and the word "data" can be replaced with "information."

[0096] [Aspects and Effects of This Embodiment] Next, an embodiment of this model and the effects of each embodiment will be described. Note that the embodiments described below are examples as of the time of filing, and this embodiment is not limited to the embodiments described below. In other words, this embodiment is not limited to the embodiments described below, and may be realized by appropriately combining the parts described above. Furthermore, lower-level embodiments may be referenced in any of the higher-level embodiments. Furthermore, the effects of this embodiment described below are merely examples, and the effects achieved by each embodiment are not limited to those described below. Also, each embodiment may achieve, for example, at least one of the effects described below.

[0097] (Aspect 1) One embodiment of an image processing apparatus is an image processing apparatus that converts input image data input to an input device into output image data within a set color gamut on an output device, comprising: a simulation image generation unit that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion unit that converts the simulation image generated by the simulation image generation unit into an XYZ color system; a confusion color information storage unit that stores confusion color information relating to confusion color lines that show the distribution of colors that are difficult for color-blind persons to distinguish; and a unit that classifies multiple similar colors into a representative number of colors. The system comprises a clustering processing unit that performs aggregated clustering, a color correction unit that uses the mixed color information of the mixed color information storage unit to perform color correction by moving the colors of the XYZ color system image data obtained by the input color information conversion unit to a second mixed color line that is spaced apart from a first mixed color line to which one of the clusters obtained by clustering is located, on a straight line connecting multiple clusters obtained by clustering, without matching the mixed color line which is a straight line formed from points where multiple mixed colors are plotted in the XYZ color system, and an output color information conversion unit that converts the image data obtained by the color correction unit into the color system of the output device. This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering.

[0098] (Aspect 2) One embodiment of the image processing apparatus is an image processing apparatus that converts input image data input to an input device into output image data within a set color gamut on an output device, comprising: a simulation image generation unit that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion unit that converts the simulation image generated by the simulation image generation unit into a Luv color system; a clustering processing unit that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction unit that performs color correction by moving the colors of the Luv color system image data obtained by the input color information conversion unit onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to a cluster, on a straight line connecting multiple clusters obtained by clustering; and an output color information conversion unit that converts the image data obtained by the color correction unit into the color system of the output device. This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering.

[0099] (Aspect 3) In one embodiment of an image processing apparatus, in the image processing apparatus according to the first or second embodiment, the image data in the input device may be color-coded to be easily distinguishable by a person with color vision deficiency. This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering.

[0100] (Aspect 4) In one embodiment of the image processing apparatus, dichromacy simulation may be performed by reducing the RGB values, converting the RGB values ​​to LMS, interpolating the L deficiency with MS, and converting the LMS values ​​back to RGB. This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering.

[0101] (Appendix 5) In one embodiment of the image processing apparatus, there may be a rotational image transfer unit that performs a rotational mapping on the XYZ color system image data obtained by the input color information conversion unit, with respect to the concentration points of the confused color lines, and a parallel mapping unit that performs a horizontal mapping on the confused color lines. This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering.

[0102] (Aspect 6) In one embodiment of the image processing apparatus, the amount of image in the re-transferred image may be determined by the ratio of the color gamut of the input device to the color gamut of the output device. This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering.

[0103] (Aspect 7) The image processing method is an image processing method that converts input image data input to an input device into output image data within a set color gamut on an output device, and comprises: a simulation image generation step of generating a color vision simulation image by performing a dichromacy simulation on the input image data in the input device; an input color information conversion step of converting the simulation image generated in the simulation image generation step into an XYZ color system; a confusion color information storage step of storing confusion color information relating to confusion color lines that show the distribution of colors that are difficult for colorblind persons to distinguish; a clustering processing step of performing clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction step of performing color correction by using the confusion color information from the confusion color information storage step and moving one cluster onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to the cluster on a straight line connecting multiple clusters obtained by clustering, without coinciding with the mixing color line which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system obtained in the input color information conversion step; and an output color information conversion step of converting the image data obtained in the color correction step into the color system of the output device. This allows image processing methods, through simulation and clustering, to enable even individuals with color vision deficiency to distinguish between color differences.

[0104] (Pattern 8) The image processing method is an image processing method that converts input image data input to an input device into output image data within a set color gamut on an output device, and comprises: a simulation image generation step of generating a color vision simulation image by performing a dichromacy simulation on the input image data in the input device; an input color information conversion step of converting the simulation image generated in the simulation image generation step into a Luv color system; a clustering processing step of performing clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction step of performing color correction by moving the colors of the Luv color system image data obtained in the input color information conversion unit onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to a cluster, on a straight line connecting multiple clusters obtained by clustering; and an output color information conversion step of converting the image data obtained in the color correction step into the color system of the output device. This allows image processing methods, through simulation and clustering, to enable even individuals with color vision deficiency to distinguish between color differences.

[0105] (Aspect 9) The image processing program is a program that causes a computer to execute an image processing function that converts input image data input to an input device into output image data within a set color gamut on an output device, and includes: a simulation image generation function that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion function that converts the simulation image generated by the simulation image generation function into an XYZ color system; a confusion color information storage function that stores confusion color information relating to confusion color lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish; and multiple similar The system performs the following functions: a clustering function that classifies colors and aggregates them into a representative number of colors; a color correction function that uses the mixed color information stored by the mixed color information storage function to correct colors by moving a cluster to a second mixed color line that is separated from the first mixed color line to which one of the clusters obtained by clustering is located, on a straight line connecting multiple clusters obtained by clustering, without making it coincide with the mixed color line, which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system obtained by the input color information conversion function; and an output color information conversion function that converts the image data obtained by the color correction function into the color system of the output device. This allows image processing methods, through simulation and clustering, to enable even individuals with color vision deficiency to distinguish between color differences.

[0106] (Aspect 10) The image processing program is a program that causes a computer to execute an image processing function that converts input image data input to an input device into output image data within a set color gamut on an output device, and executes the following functions: a simulation image generation function that performs a dichromacy simulation on the input image data in the input device to generate a color vision simulation image; an input color information conversion function that converts the simulation image generated by the simulation image generation unit into a Luv color system; a clustering processing function that performs clustering to classify multiple similar colors and aggregate them into a representative number of colors; a color correction function that performs color correction by moving the colors of the Luv color system image data obtained by the input color information conversion function onto a second confusion color line that is separated from a first confusion color line where one cluster coincides or is close to a cluster on a straight line connecting multiple clusters obtained by clustering; and an output color information conversion function that converts the image data obtained by the color correction function into the color system of the output device. This allows image processing methods, through simulation and clustering, to enable even individuals with color vision deficiency to distinguish between color differences.

[0107] This allows the image processing device to distinguish color differences even for people with color vision deficiency through simulation and clustering. [Explanation of symbols]

[0108] 1 Image processing device 2 keyboards 3 monitors 4 Printers 5 Input / Output Interfaces 6 CPU 7 RAM 8 Program ROM 9 HDD 10 Simulation Image Generation Unit 20 Input color information conversion unit 30. Color Confusion Information Storage Unit 40 Clustering Processing Unit 50 color correction section 51st Transfer Image Section 52 Parallel mapping section 60 Output color information conversion unit

Claims

1. An image processing device that converts input image data input to an input device into output image data within a color gamut set on an output device, The input device includes a simulation image generation unit that performs a dichromacy simulation on the input image data to generate a color vision simulation image, The aforementioned simulation image generation unit converts the simulation image generated by the simulation image generation unit into an XYZ color system, A color confusion information storage unit stores color confusion information related to color confusion lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish, A clustering processing unit that classifies multiple similar colors and aggregates them into a representative number of colors, A color correction unit performs color correction by using the mixed color information of the mixed color information storage unit to move the color of the XYZ color system image data obtained by the input color information conversion unit to a second mixed color line that is spaced apart from a first mixed color line where one cluster coincides or is close to a cluster, on a straight line connecting multiple clusters obtained by the clustering process, without matching it to a mixed color line which is a straight line formed from points where multiple mixed colors in the XYZ color system are plotted. An output color information conversion unit converts the image data obtained by the color correction unit into the color system of the output device, An image processing apparatus characterized by having

2. An image processing device that converts input image data input to an input device into output image data within a color gamut set on an output device, The input device includes a simulation image generation unit that performs a dichromacy simulation on the input image data to generate a color vision simulation image, The aforementioned simulation image generation unit converts the simulation image generated by the simulation image generation unit into a Luv color system, A clustering processing unit that classifies multiple similar colors and aggregates them into a representative number of colors, A color correction unit performs color correction by moving the color of the Luv color system image data obtained by the input color information conversion unit onto a second confused color line that is spaced apart from a first confused color line where one cluster coincides with or is close to a cluster, on a straight line connecting multiple clusters obtained by the clustering. An output color information conversion unit converts the image data obtained by the color correction unit into the color system of the output device, An image processing apparatus characterized by having

3. The image processing apparatus according to claim 1 or 2, characterized in that the image data in the input device is color-coded in a way that makes it easy for a person with color vision deficiency to identify.

4. The image processing apparatus according to claim 1 or 2, characterized in that the dichromacy simulation is performed by reducing the RGB values, converting the RGB values ​​to LMS, interpolating the L loss with MS, and converting the LMS values ​​back to RGB.

5. A rotational transfer image unit performs a rotational mapping on the XYZ color system image data obtained by the input color information conversion unit, with the confusion line concentration point being the center of the confusion line concentration point. A parallel mapping unit that performs a horizontal mapping on the aforementioned color confusion lines, The image processing apparatus according to claim 1, characterized by having the following features.

6. The image processing apparatus according to claim 5, characterized in that the amount of image in the aforementioned transferred image is determined by the ratio of the color gamut of the input device to the color gamut of the output device.

7. In an image processing method that converts input image data input to an input device into output image data within a color gamut set on an output device, A simulation image generation step of generating a color vision simulation image by performing a dichromacy simulation on the input image data in the input device, An input color information conversion step that converts the simulation image generated in the simulation image generation step into an XYZ color system, A step of storing confused color information, which stores confused color information related to confused color lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish, A clustering process step that classifies multiple similar colors and aggregates them into a representative number of colors, A color correction step is performed by using the mixed color information obtained in the mixed color information storage step and moving the clusters obtained by the clustering to a second mixed color line that is spaced apart from a first mixed color line to which one of the clusters coincides or is close, without making it coincide with the mixed color line which is a straight line formed from points plotted by multiple mixed colors in the XYZ color system obtained in the input color information conversion step. An output color information conversion step which converts the image data obtained in the color correction step into the color system of the output device, An image processing method characterized by having the following features.

8. In an image processing method that converts input image data input to an input device into output image data within a color gamut set on an output device, A simulation image generation step of generating a color vision simulation image by performing a dichromacy simulation on the input image data in the input device, The above simulation image generation step includes an input color information conversion step that converts the generated simulation image into a Luv color system, A clustering process step that classifies multiple similar colors and aggregates them into a representative number of colors, A color correction step is performed by moving the color of the Luv color system image data obtained in the input color information conversion step onto a second confused color line that is spaced apart from a first confused color line where one cluster coincides with or is close to a plurality of clusters obtained by the clustering, on a straight line connecting the clusters. An output color information conversion step which converts the image data obtained in the color correction step into the color system of the output device, An image processing method characterized by having the following features.

9. In a program that causes a computer to perform an image processing function that converts input image data received on the input device side into output image data within the color gamut set on the output device side, The input device includes a simulation image generation function that performs a dichromacy simulation on the input image data to generate a color vision simulation image, An input color information conversion function that converts the simulation image generated by the simulation image generation function into an XYZ color system, A function for storing confused color information related to confused color lines that show the distribution of colors that are difficult for people with color vision deficiency to distinguish, A clustering processing function that classifies multiple similar colors and aggregates them into a representative number of colors, A color correction function that uses the mixed color information stored by the mixed color information storage function to correct colors by moving a cluster to a second mixed color line that is spaced apart from a first mixed color line where one cluster coincides with or is close to a cluster, without making it coincide with the mixed color line, which is a straight line formed from points plotted by the XYZ color system obtained by the input color information conversion function. An output color information conversion function that converts the image data obtained by the color correction function into the color system of the output device, An image processing program characterized by causing the following to be executed.

10. In a program that causes a computer to perform an image processing function that converts input image data received on the input device side into output image data within the color gamut set on the output device side, The input device includes a simulation image generation function that performs a dichromacy simulation on the input image data to generate a color vision simulation image, An input color information conversion function that converts the simulation image generated by the simulation image generation function into the Luv color system, A clustering processing function that classifies multiple similar colors and aggregates them into a representative number of colors, A color correction function that corrects the color of Luv color system image data obtained by the input color information conversion function by moving the color of the image data obtained by the clustering function onto a second confused color line that is spaced apart from a first confused color line where one cluster coincides with or is close to another cluster, on a straight line connecting multiple clusters obtained by the clustering function. An output color information conversion function that converts the image data obtained by the color correction function into the color system of the output device, An image processing program characterized by causing the following to be executed.

Images