Color correction method and apparatus

Abstract

To allow a minority color vision person to obtain color vision coincident with that of many standard color vision persons by coping with an optical characteristic of a display device and an individual difference in color vision.SOLUTION: A pre-conversion image 100 is displayed on a display 130 for a majority color vision person via a normal display device 120 on one hand, and is displayed on a display 170 for a minority color vision person via a color vision correction device 150 on the other hand. In color vision correction device 150, correction calculator 160 performs correction conversion calculation on the RGB data (R, G, B) according to correction conversion calculation program 162 to obtain corrected RGB data (R ', G ', B '). At this time, the correction conversion is performed so that the cone stimulation values (L, M, S) for a person with polychromatism who views the display 130 and the cone stimulation values (L ', M ', S ') for a person with minority color vision who views the display 170 match.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an improvement in a color correction method and device for correcting the color of an image in accordance with human color perception (color vision). [Background technology]

[0002] For people with color vision that differs from the majority of people's color vision (majority or standard color vision), this is often a congenital or genetic trait, and there is currently no cure. As a result, they are unable to experience and share the excitement of paintings, anime, photographs, movies, etc. in the same way as people with majority color vision. Moreover, they will lose such aesthetic opportunities throughout their lives.

[0003] Meanwhile, in everyday life, various displays using color are widely used. For example, traffic signs, station guides, and electrical appliances convey their meanings through color identification information. However, these color displays are intended to be perceived by color minorities. Therefore, depending on the color used, the meaning of the display may not be accurately conveyed and may be misinterpreted by color minorities. Therefore, in order to ensure that even color minorities can easily recognize the displayed content, a concept called Color Universal Design (CUD) has been proposed, which limits the colors used to those easily distinguishable by color minorities. However, this approach limits the colors available to both color minorities and color majorities, and color minorities do not subjectively perceive colors in the same way as color minorities.

[0004] Even when limiting the colors of CUD to those that are easy to recognize, or replacing them with colors that are easy to recognize, or replacing them with other information such as letters or grids, the colors that can be perceived by those with major color vision and those with minor color vision, or the relationships between colors, do not match. Therefore, these methods cannot be said to be effective aids, especially in fields such as art, illustration, design, and video content, where the color scheme effect is strong and the communication of the content is important.

[0005] In recent years, there has been an increase in the number of devices capable of accurately displaying colors, as well as the digital content, such as images, that can be displayed on them. Furthermore, research has shown that approximately one in 20 men has some type of color vision minority, so it is believed that many color vision minorities would benefit if they were able to perceive color schemes in the same way as color vision majorities.

[0006] Background art for color vision correction or color correction is, for example, the "Display Device, Display Control Device, and Display Method" described in Patent Document 1 below. This aims to enable people with minority color vision to indirectly experience hues similar to those of people with majority color vision, and generates corrected RGB data from color indication value R (red), G (green), and B (blue) data using a correction matrix in equation 1 of the publication. This correction matrix is ​​the inverse matrix of the Daltonize matrix. Non-Patent Document 1 will be described later. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-57621 [Non-patent literature]

[0008] [Non-Patent Document 1] Journal of the Institute of Image Electronics Engineers, Vol. 35, No. 4 (2006) "Extended Color Space and Color Management Technology I, 1.1 Overview of Color Management Technology" Summary of the Invention [Problem to be solved by the invention]

[0009] Incidentally, most cases of minority color vision are caused by a shift in the cone sensitivity distribution in the wavelength direction. This shift can be diverse and even mixed, and exists almost continuously. However, the Daltonize matrix in the background art described above does not take into account the characteristics of the optical elements in the display device or the cone sensitivity distribution of individuals with minority color vision, and therefore cannot match the color perception of majority and minority color vision individuals. Therefore, the performance of correction transformations based on the Daltonize matrix is ​​limited to allowing majority and minority color vision individuals to experience similar hues.

[0010] The present invention was made in consideration of these points, and aims to enable people with minority color vision to obtain color perception (hereinafter referred to as "color vision") or a color vision experience that is consistent with that of most people with standard color vision, while also addressing the optical characteristics of display devices and individual differences in color vision. [Means for solving the problem]

[0011] The present invention is characterized in that, when correcting image data of an image before correction that is to be displayed to a person with minority color vision to obtain image data of an image after correction that is to be displayed to a person with minority color vision, the image data of the image before correction is corrected so that the cone stimulation values ​​when the person with majority color vision views the image before correction match the cone stimulation values ​​when the person with minority color vision views the image after correction to obtain the image data of the image after correction.

[0012] According to one main aspect, the correction of the image data is calculated based on the cone sensitivity distribution of a color minority and the spectral luminance distribution of the image display means of the image data. According to another aspect, the image data before correction is converted into CIE XYZ image data, and the correction is performed on the converted image data. Furthermore, the CIE XYZ image data is further converted into CIE Lab image data, and the converted CIE Lab image data is adjusted so that the corrected image can be displayed on the display means. According to yet another aspect, the color reproduction gamut for a color minority and the color reproduction gamut for a color majority are displayed supplementarily in the same color space. The above and other objects, features, and advantages of the present invention will become apparent from the following detailed description and the accompanying drawings. [Effects of the Invention]

[0013] According to the present invention, the video signal is corrected so that the cone sensitivity distribution is consistent between polychromats and minority color vision persons, so that color correction can be performed that is easily recognizable by minority color vision persons and also accommodates individual differences in color vision. [Brief explanation of the drawings]

[0014] [Figure 1] This provides an overview of how humans perceive color. [Figure 2] 1 shows the configuration of devices according to first and fifth embodiments of the present invention. [Figure 3] 1 shows an example of the spectral luminance distribution and the cone sensitivity distribution of a display. [Figure 4] 1 shows the relationship that the general-purpose correction conversion in the color vision correction device in the first embodiment must satisfy. [Figure 5] The relationship that the correction transformation in the second embodiment must satisfy is shown below. [Figure 6] A third embodiment will be described in which an adjustment unit for displaying all the colors of the original data is incorporated into the correction calculation. [Figure 7] 10 shows the relationship of the area in the color space that the adjustment unit should satisfy in the third embodiment. [Figure 8]A fourth embodiment of a video data display system using a color vision correction device will be described. [Figure 9] 5 shows an example 5 of an auxiliary display relating to correction transformation that is useful to users. DETAILED DESCRIPTION OF THE INVENTION

[0015] Hereinafter, embodiments of the present invention will be described in detail based on examples. A,(·) indicates a list of vector or matrix elements or numerical values. b,[·] indicates a vector or a set of numbers represented as a symbol. c,Double brackets [[·]] are used to represent matrices symbolically. d,func(·) represents a function that takes · as an argument, and the character string immediately before (·) varies depending on the function. e, Alphabetical characters represent numbers is. [Example]

[0016] First, referring to Figure 1, we will provide an overview of the mechanism by which humans sense color. Figure 1 is an outline of how color vision works in humans, showing that when light enters the eyeball EY, the photoreceptor cells PC in the retina are stimulated, and a signal according to the magnitude of each stimulus is transmitted to the brain BR via the optic nerve (not shown). There are three types of photoreceptor cells PC that provide the sense of color: S cones CS, M cones CM, and L cones CL. (1) L cone CL: Light near long wavelengths (2) M cone CM: Light with medium wavelengths (3) S cone CS: Light with short wavelengths There are also photoreceptor cells called rods (not shown). The wavelength sensitivity characteristics of these three types of cones CL, CM, and CS are shown in Figure 3(B).

[0017] Signals from these cones CL, CM, and CS, corresponding to the magnitude of the stimulus, are transmitted via the optic nerve to the brain BR. The brain BR is said to perceive "color" based on the combination of the signal intensities of the cones CL, CM, and CS. For example, lightness XL is perceived as the approximate sum of the signals from the M cones CM and L cones CL. Blue and yellow are perceived as the blue-yellow signal Cb-y, which is the approximate difference between the yellow signal Cy obtained from the approximate sum of the signals from the M cones CM and L cones CL and the signal from the S cones CS. Red and green are perceived as the red signal Cr obtained from the approximate sum of the signals from the S cones CS and L cones CL, and the red-green signal Cr-g, which is the approximate difference between the signal from the M cones CM.

[0018] The color perceived by the brain is thought to be a combination of three values: lightness (XL), blue-yellow signal (Cb-y), and red-green signal (Cr-g). These three values ​​indicate brightness: lightness (XL), blue-yellow signal (Cb-y), and red-green signal (Cr-g). Treating these three values ​​as Cartesian coordinates creates a color space that is easy to interpret for color relationships. Therefore, if these three values ​​are consistent between polychromats and minorities, the relationships between colors—that is, the perceived effects of the color scheme—are expected to be consistent. For example, a display using only colors with the same brightness for polychromats can be presented as a color scheme with the same brightness for minorities.

[0019] It is believed that minority color vision occurs because the cone sensitivity distribution deviates from the above-mentioned Figure 3(B). Therefore, if the strength of the cone CL, CM, and CS signals generated in minority color vision persons is the same as the strength of the cone CL, CM, and CS signals generated in majority color vision persons, it is thought that majority color vision persons and minority color vision persons will be able to perceive the same colors.

[0020] This is shown in Figure 2, where a pre-conversion image 100 is displayed on a display 130 for color vision sufferers via a normal display device 120, and on a display 170 for color vision sufferers via a color vision correction device 150. A wide color gamut display with good color reproducibility is preferable as the display 170 for color vision sufferers. In the normal display device 120, the pre-conversion image is stored as RGB data (R, G, B) in the image memory 122. In the color vision correction device 150, the pre-conversion image is stored as RGB data (R, G, B) in the image memory 152. In Figure 2, the adjustment calculation program 164 of the correction calculation unit 160, the auxiliary display calculation unit 400, and the auxiliary display calculation program 402 will be described later.

[0021] In the color vision corrector 150, a correction calculation unit 160 performs correction conversion calculations on the RGB data (R, G, B) stored in the video memory 152 using a correction conversion calculation program 162, and the resulting data is stored in the video memory 152 as corrected RGB data (R', G', B'). In a normal display device 120, the RGB data (R, G, B) in the video memory 122 is output directly to the display 130. On the other hand, in the color vision corrector 150, the corrected RGB data (R', G', B') is output to the display 170.

[0022] When a person with majority color vision views display 130, they see an image with normal colors, but when they view display 170, they see an image with color bias. Conversely, when a person with minority color vision views display 130, they see an image with color bias, but when they view display 170, they see an image without color bias. In other words, when a person with minority color vision views display 170, they can perceive the same colors as a person with majority color vision views display 130.

[0023] However, in either case, the degree of stimulation of the cones CL, CM, and CS of the observer's photoreceptor cells PC is the same. Here, if the cone stimulation values ​​of a polychromat who observes the image on display 130 are (L, M, S), and the cone stimulation values ​​of a polychromat who observes the image on display 170 are (L', M', S'), then the cone stimulation values ​​(L, M, S) and (L', M', S') will be the same. That is, (R,G,B) ≠ (R',G',B') (L,M,S)=(L',M',S') The following relationship holds. In this way, by correcting the corrected RGB data (R', G', B') so that the cone stimulation values ​​match between the majority and minority color vision individuals and outputting it to the display 170 for minority color vision individuals, the majority and minority color vision individuals can perceive similar colors. Note that although FIG. 2 above has been described as if the images on the displays 130 and 170 were viewed simultaneously by the majority and minority color vision individuals, this does not have to be the case.

[0024] In the case of inverse transformation, the RGB data (R, G, B) is corrected so that the cone stimulation values ​​match between the majority and minority color vision persons, and output to the display 130 for the majority color vision person. By observing the image, the majority color vision person can experience colors similar to the colors of the image experienced by the minority color vision person.

[0025] Next, a method for calculating the correction conversion of RGB data will be described with reference to Figs. 3 and 4. First, the relationship between color indication values ​​(R, G, B) and cone stimulation values ​​(L, M, S) using the RGB color space will be shown. The spectral luminance distribution of a display is, for example, as shown in Fig. 3(A), while the distribution of human cone sensitivity is as shown in Fig. 3(B). Here, the relationship between the color indication values ​​(R, G, B) of a display and the cone stimulation values ​​(L, M, S) of cones CL, CM, and CS is expressed by the following equation (1) with respect to wavelength λ. Note that the RGB values ​​are those after the display gamma has been applied.

number

[0026] In the formula, "r(λ),g(λ),b(λ)" represents the spectral luminance distribution in Figure 3(A), and "l(λ),m(λ),s(λ)" represents the cone sensitivity distribution of the majority of color vision persons. In these formulas, when the spectral luminance distribution "r(λ),g(λ),b(λ)" and the cone sensitivity distribution "l(λ),m(λ),s(λ)" are integrated, it is shown that the color indication values ​​(R,G,B) and the cone stimulation values ​​(L,M,S) have the matrix relationship shown in Equation 2 below.

number

[0027] When calculating the transformation matrix in Equation 2, for example, the spectral luminance distribution of the display (r(λ), g(λ), b(λ)) can be measured using a spectrophotometer, and the cone sensitivity distribution of the majority color vision person (l(λ), m(λ), s(λ)) can be calculated using the values ​​estimated by Smith and Pokorny. Similarly, for minority color vision persons, the relationship between the color indicators (R', G', B') and the cone stimulation values ​​(L', M', S') can be calculated using the measured spectral sensitivity distribution and the estimated cone sensitivity distribution. The cone stimulation values ​​calculated as above when a majority color vision person views a display and the cone stimulation values ​​when a minority color vision person views a display are matched, as shown in Figure 4. In other words, by correcting the color indicators (R, G, B) of the display to (R', G', B') so that (L, M, S) = (L', M', S'), the minority color vision person can view the same image as the majority color vision person and experience the same color vision. This relationship is shown in equation 3, where [[C]] represents the correction transformation matrix.

number

[0028] As described above, according to this embodiment, the color indication values ​​are corrected so that the cone stimulation values ​​corresponding to any color indication value (for example, (R, G, B)) are consistent between major and minority color vision persons, so that the perceived colors and the relationships between colors are absolutely consistent between color vision persons. This is extremely useful for minority color vision persons who view video content in which color schemes are important, and for designers, illustrators, video creators, and other color vision persons who require advanced color calibration. [Example]

[0029] Next, an example of a specific method for converting and correcting color indication values ​​(R, G, B) to (R', G', B') will be described with reference to FIG. [R] = (R, G, B) indicates the color value in the RGB color space, [X]=(X,Y,Z) represents the color value in the CIE XYZ color space, [Lab] = (L, a, b) indicates the color value in the CIE Lab color space, [L] = (L,M,S) indicates the cone stimulation value. In the above formula, the left side indicates a vector, and the right side indicates a set of numerical values ​​of its elements. The subscript "l" indicates low color gamut, "h" indicates high color gamut, and the subscript "n" indicates majority color vision and "v" indicates minority color vision. For example, the cone stimulation value (L', M', S') for a minority color vision person is expressed as the vector [L V ].

[0030] The RGB color space is a representation of color in which perceived colors vary depending on the color profile (a configuration file for managing color representation) used. In contrast, the CIE XYZ color space is used as an absolute representation of color that uniquely indicates perceived colors among people with multiple color vision. On the other hand, the CIE Lab color space is a display method that emphasizes expressing psychological color relationships. Its coordinate axes are L, which represents lightness, and a and b, which represent the relationship between complementary colors. When the lightness L is constant, the brightness will be the same regardless of which complementary colors a and b are selected, and when the angle between complementary colors a and b is fixed, the hue will be the same. Therefore, this color space makes it easy to manipulate the psychological color representation quantities of hue, saturation, and lightness.

[0031] The transformation matrix and functions represent the following, respectively. (1) The matrix [[X]] is a matrix for converting color indication values ​​from the RGB color space to the CIE XYZ color space, and satisfies the relationship [X] = [[X]][R] for the above vectors [X] and [R]. The elements of the matrix [[X]] can be obtained from the color profile of the CMS (Color Management System, see, for example, Non-Patent Document 1).

[0032] (2) The matrix [[L]] is used to convert from the RGB color space to cone-stimulus values, and satisfies the relationship [L] = [[L]][R] for the above vectors [L] and [R]. The matrix elements of the matrix [[L]] are calculated using the above formula 1.

[0033] (3) The correction transformation matrix [[C]] represents a correction transformation matrix of color indicator values ​​that allows a minority color vision person to perceive the same colors as a majority color vision person, as described above. The color indicator values ​​to be converted may be any color indicator values, including those in the examples described above, but in this embodiment, the above-mentioned [X] is selected. This eliminates the need to modify the correction transformation matrix [[C]] once it has been calculated, even if the original RGB data is changed to different RGB data defined in a different color space.

[0034] Furthermore, the CIE XYZ color space is a color specification method that allows multiple color vision users to absolutely specify the color they perceive (this corresponds to PCS in the color management system. See Non-Patent Document 1, 3.1 Color Management Basic System). However, by using the vector [X] as the target of conversion with the correction transformation matrix [[C]], the range of color vision that can be specified absolutely with the vector [X] before conversion can be expanded to include not only standard color vision users but also those with minority color vision users. In other words, in this example, the same color can be imagined for a certain vector [X] regardless of color vision, making it easier to specify colors across minority color vision users. However, in practice, the effect of display gamma must also be taken into account.

[0035] In this embodiment, the correction transformation is not a function but [X v It is expressed as a correction transformation matrix [[C]] that satisfies the relationship ]=[[C]][X]. By calculating this correction transformation matrix [[C]] in advance, it becomes possible to correct colors in real time.

[0036] (4) cyl(·) represents a function that converts color indication values ​​(intensities of three light sources) based on three stimuli, such as vectors [R] and [X], into indication values ​​in a color space that can be considered cylindrical, with psychological hue, saturation, and brightness as its coordinate axes. In this embodiment, the relationship [Lab] = cyl([X]) is satisfied, converting from the CIE XYZ color space, which can absolutely indicate colors perceived by a large number of color vision scientist, to the CIE Lab color space, which is known to accurately express psychological color characteristics. However, the white point used in the conversion is the white point of the RGB data of the image to be displayed, i.e., the white point in the low-gamut RGB color space.

[0037] (5) ajst(·) is a function that adjusts color values ​​so that the corrected color values ​​are within the display range of the wide color gamut display, and satisfies the relationship of the formula [Lab'] = ajst([Lab]). For example, the color mapping technique described in "4.4 Mapping Method" in the aforementioned Non-Patent Document 1 may be used as this adjustment function ajst(·). In this embodiment, adjustment is performed in the CIE Lab color space before performing correction conversion using the correction conversion matrix [[C]].

[0038] As mentioned above, the CIE XYZ color space before the correction transformation using the correction transformation matrix [[C]] extends the range of color vision that can be absolutely specified to include not only standard color vision but also minority color vision. Therefore, the relationships between colors in the CIE Lab color space converted from the CIE XYZ color space are the same for majority color vision individuals and minority color vision individuals. Therefore, color manipulation, such as changing saturation to maintain constant lightness, is possible for minority color vision individuals just as it is for majority color vision individuals. As a result, the effects of color mapping described in Non-Patent Document 1, i.e., psychological lightness preservation, saturation preservation, color difference minimization, etc., can be achieved for minority color vision individuals just as they are for majority color vision individuals.

[0039] This embodiment is divided into a phase for calculating the correction transformation matrix [[C]] and a phase for calculating the adjustment function ajst(·) on the calculated correction transformation matrix [[C]]. Therefore, the phase for calculating the correction transformation matrix [[C]] will be explained first. Note that, for the purposes of calculation, the cone stimulus values ​​(L, M, S) that are assumed to be received by a majority color vision person are hypothetical, so any display can be used. In this embodiment, it is assumed to be the same as the wide color gamut display used by a minority color vision person. The correction transformation matrix [[C]] can be obtained by solving the relationship between the three values, such as vectors, as an equation in Figure 5. While the process of solving the equation itself is omitted, the equations that hold true will be explained step by step with reference to Figure 5.

[0040] First, calculate the cone stimulation values ​​(L, M, S) that a hypothetical majority of color vision persons would receive. (1) Steps S10 and S12: First, the low color gamut RGB data [R l ] is converted into a vector [X] that indicates the color indication value data in the CIE XYZ color space, which is a uniform color specification for all color vision.

[0041] (2) Step S14: Next, the vector [X] indicating the data of the color indication values ​​in the CIE XYZ color space obtained as described above is converted into a transformation matrix [[X h ]]'s inverse matrix [[X h ]] -1 , and the vector [R h n ].

[0042] (3) Step S16: Next, a vector [R h n ], after gamma processing, the matrix [[L h n ]] to obtain the vector [L n ] to

[0043] Next, the cone stimulation values ​​(L', M', S') received by a color minority using a correction device are calculated as follows: This calculation process includes the desired correction transformation matrix [[C]]. (1) Step S20: The color indication value data [X] in the CIE XYZ color space is converted into a vector [X] indicating the color indication value data after the correction conversion using a correction conversion matrix [[C]] that converts the color indication value so that a minority color vision person can perceive the same color as a majority color vision person. V ] to

[0044] (2) Step S22: The vector [X V ] into the transformation matrix [[X h ]]'s inverse matrix [[X h ]] -1, and the vector [R h v ].

[0045] (3) Step S24: Next, a vector [R h v ] is processed by gamma processing, and then the matrix [[L h V ]], the vector of cone stimulation values ​​for the minority color vision person [L V ] to

[0046] In this case, the cone stimulation value [L n ] and the cone stimulation value in minority color vision [L v ] is required to match, a correction transformation matrix [[C]] that satisfies this requirement can be solved. By using the obtained correction transformation matrix [[C]], when RGB data of any color gamut or color indication values ​​in the CIE XYZ color space are given, the corrected RGB data to be input to the wide color gamut display to be used can be obtained.

[0047] In this way, according to this embodiment, the image data before correction is converted into CIE XYZ image data, and the above correction is performed on the converted image data. By doing so, compared to the case where the correction is performed on the RGB image data, a. Even if the low color gamut RGB data to be displayed becomes RGB data defined in a different color space, there is no need to modify the correction transformation matrix [[C]] that has already been calculated. b) In addition, in the vector [X] of color indication values ​​in the CIE XYZ color space, the range of color vision that can be absolutely indicated is extended to include not only major color vision but also minor color vision. Such technical effects can be obtained. [Example]

[0048] Next, a third embodiment of the present invention will be described with reference to FIGS. 6 and 7. When the correction conversion is performed in the second embodiment, the RGB data corrected by the matrix [[C]] described above may take values ​​in a range that the display device cannot accept. For example, if the input range of RGB values ​​of a display device is only between "0" and "1," negative values ​​(less than "0") or values ​​greater than "1" would be included. Therefore, in this embodiment, an adjustment is made to naturally round the RGB data input to a wide color gamut display to within a range that the display device can accept (display). This adjustment is performed by the adjustment calculation program 164 of the correction calculation unit 160 shown in FIG. 2. In this embodiment, the adjustment function ajst(·) is first calculated. The adjustment function ajst(·) is incorporated into the correction conversion process as shown in FIG. 6. Note that in FIG. 6, the same reference numerals are used for the same components as those in FIG. 5, and their description will be omitted.

[0049] The adjustment using the adjustment function ajst(·) is performed as follows, as shown in steps S30 to S34 of FIG. (1) Step S30: Cylindrical coordinate system transformation for a vector [X] of color indication values ​​in the CIE XYZ color space, expressed as [Lab] = cyl([X]), (2) Step S32: Adjustment of [Lab] by the adjustment function ajst(·) as [Lab'] = ajst([Lab]), (3) Step S34: cyl -1 Inverse transformation to [X'] by (·), Then, [X'] is corrected by the correction transformation matrix [[C]].

[0050] In this way, because the color indication values ​​are adjusted at a stage prior to conversion using the correction transformation matrix [[C]], the range (hereafter indicated by {·}) that the color indication values ​​[Lab'] can take in the adjusted CIE Lab color space must correspond to the range that the wide color gamut display being used can display in the CIE XYZ space after conversion using the correction transformation matrix [[C]].

[0051] Next, each step will be described with reference to Fig. 7. Fig. 7 shows the procedure for providing the color range to be taken before and after adjustment, which is required for compressing the color gamut using the adjustment function ajst(·).

[0052] (1) Steps SA to SC: The area in the CIE XYZ space that can be displayed by the wide color gamut display is defined as {X h}, then by converting using the inverse matrix of the correction transformation matrix [[C]], the area {X c-1h}.

[0053] (2) Step SD: Next, the obtained region {X c-1h} into the CIE Lab color space, the cylindrical coordinate system transformation function cyl(·) is used to transform the CIE Lab color space into the area {Lab c-1h} is obtained. The same applies to the RGB data with a small color gamut to be corrected (see steps SE to SG).

[0054] As a result, the adjustment function ajst(·) converts the color indication value [Lab] in the area {Lab1} on the CIE Lab color space that the low color gamut RGB data to be corrected can take into the area {Lab c-1h It can be seen that any adjustment that fits into the {} is sufficient. In order for this processing to be a natural adjustment, it is desirable that the conversion processing take into account the psychological structure of color, and the above-mentioned color mapping (see "4.4 Mapping Method" in Non-Patent Document 1) can also be used.

[0055] In this way, by further converting the CIE XYZ image data into CIE Lab image data, adjusting it, and then performing the correction conversion, the technical effect is achieved that the RGB data can be rounded into a range that can be received (displayed) by a display device through an adjustment process that reflects the psychological structure of color, even for people with minority color vision. [Example]

[0056] Next, a fourth embodiment of the present invention will be described with reference to Fig. 8. Fig. 8(A) shows an example of a system that uses the above-mentioned color vision corrector 150 to perform various displays for color vision minorities. A personal computer 200, a camera or video camera 202, and a scanner 204 are connected to the input side of the image data signal of the color vision corrector 150, and the color indication values ​​(R, G, B) of the image data are respectively input to the color vision corrector 150. For example, the color indication values ​​of the image data displayed on the screen are input from the personal computer 200, the color indication values ​​of the image data that has been photographed are input from the camera 202, and the color indication values ​​of the image data that has been read are input from the scanner 204.

[0057] The color indication values ​​(R, G, B) of these video data are subjected to color vision correction by a color vision correction device 150, and the color indication values ​​(R', G', B') of the corrected video are output to a display 170, a VR device 172 such as a head-mounted display, and a liquid crystal pen tablet 174. In other words, video data with accurate color reproduction when viewed by a color vision minority is displayed on the display 170, etc.

[0058] Recent VR devices are equipped with a color pass-through function that projects the outside world into a head-mounted display. For example, a person with color vision minority could wear the device in a museum, convert the outside world into video data, and project the color-corrected video into the head-mounted display, allowing them to share the same color vision experience as people with color vision majority who are viewing the artwork around them.

[0059] If the personal computer 200 is equipped with a video / video content production application, a person with minority color vision can always view and edit the corrected and converted video on the display 170. For example, the HSL / HSV color picker, which previously did not appear uniform, now appears to have uniform color changes, and reference materials displayed on the display 170 are also displayed with the color balance seen by a person with majority color vision. In particular, the balance of brightness and saturation intensity can be severe, so when creating a work collaboratively, it is useful to share this sense of color balance through materials.

[0060] Figure 1B shows an example in which the completed video data for a color minority is later corrected for a color majority and displayed. The color indication values ​​(R', G', B') of the video data created by a color minority using the tablet 300 are subjected to inverse color vision correction by the color vision correction device 150, and the color indication values ​​(R, G, B) of the video data are output to the display 310. The colors of the video data input by the tablet 300 are accurately reproduced by a color minority, but when viewed by a color majority, the video data appears to have shifted colors. However, by performing an inverse conversion from (R', G', B') to (R, G, B) by the color vision correction device 150, video data with accurately reproduced colors is displayed on the display 310 when viewed by a color majority. [Example]

[0061] Next, a fifth embodiment of the present invention will be described. The present invention is based on the premise that the cone sensitivity distribution of a color minority is known. It is believed that the cone sensitivity distribution can be estimated from genetic factors or the results of certain color vision tests. However, if it is difficult to know the estimated results, when the color indication values ​​of video data for a color minority are corrected for viewing by a color minority, a cross section of an appropriate color solid or the like may be displayed on the display screen, and an adjustment amount may be selected that will result in uniform color changes for the color minority. Alternatively, the amount of correction may be adjusted by referring to appropriate content to determine the amount of correction that will result in good color perception. [Example]

[0062] Next, a sixth embodiment of the present invention will be described with reference to Fig. 9. In Fig. 9(A), when using this device, the range of colors perceivable by a majority color vision person is shown as a region indicated by a solid line, and the range of colors perceivable by a minority color vision person is shown as a region indicated by a dotted line. At each point on the graph, the cone stimulation values ​​experienced by the majority color vision person and the minority color vision person are the same.

[0063] In the simplest terms, the area {Lab1} shown in step SG in Figure 7, which is before and after the adjustment function ajst(·), and the area {Labc-1h} shown in step SD, can be plotted as a comparison of areas in the CIE Lab color space, but Figure 9(A) shows this converted into the low gamut RGB color space, which is the gamut of the low gamut RGB data to be corrected. Any color space can be used, and the shape of the graph will change depending on which color space is used and the color gamut of the display used.

[0064] Furthermore, the volume of the area that can be accurately reproduced can be displayed as a percentage of the volume of the area in the color space that can be used by the color gamut of the low color gamut RGB data to be corrected. In terms of uniformity of distribution in color perception, it is useful to perform calculations in the CIE Lab color space or the CIE Luv color space. As shown in Example 2, these displays are made possible by the characteristic of the present invention that can uniformly indicate colors perceived by any color vision person. Since these displays are based on the possibility of correction in the CIE XYZ color space, they can also be converted to any other color space for display depending on the purpose.

[0065] Figure 9(B) is an auxiliary display that makes it easy to see how the color indication values ​​change when the range of colors that can provide cone stimulation for polychromats but cannot provide cone stimulation for minority color vision persons, i.e., the area outside the area indicated by the dotted line but within the area indicated by the solid line in Figure 9(A), is compressed and adjusted to fit within the area indicated by the dotted line by color mapping using a GMM (Gamut Mapping Module) as described in the above-mentioned Non-Patent Document 1 (this is generally called a rendering indent; see Non-Patent Document 1, section 4. GMM "Gamut Mapping Module").

[0066] The simplest example is a comparison of the color indicator vector [Lab] in the CIE Lab color space (step S30 in Figure 6) before and after the adjustment function ajst(·) with the color indicator vector [Lab'] in step S32. Figure 9(B) shows the color indicator values ​​converted to the low-gamut RGB color space, which is the gamut of the low-gamut RGB data to be corrected, and then converted to HSV color space, and the resulting color indicator values ​​are displayed in polar coordinates. The angular direction is hue H, and the radial direction is either saturation S or lightness V. Sets of colors with constant saturation S and lightness V for multiple color vision users are represented by solid circles in this graph. On the other hand, the color indicator values ​​obtained by color compression using the color mapping described above are represented by dotted lines. By comparing the two, it is possible to see at a glance which hues cannot be expressed with their original lightness and saturation even when using this device. Figure 9(B) shows an example of the change in lightness V, in particular, due to color compression adjustment.

[0067] Figure 9(C) is an auxiliary display that allows easy confirmation of how the color indicator values ​​for each hue H have changed due to the correction transformation matrix [[C]]. The horizontal axis represents hue H. The graph in this figure shows a set of colors corrected and transformed when saturation S and lightness V are constant for most color vision sufferers. Thin lines represent the color indicator values ​​(R, G, B) in the RGB color space, while thick lines represent the difference between the hue H, saturation S, and lightness V values ​​before and after conversion. The graph representation of the color indicator values ​​(R, G, B) in the RGB color space can also be shown as the difference between before and after conversion, just like the color indicator values ​​(H, S, V) in the HSV color space.

[0068] This display is the [X] in step S10 and the [X] in step S20 in FIG. 5 before and after the correction transformation matrix [[C]]. v ], for example, [[X h ]] -1The color is converted to the wide gamut RGB color space in the gamut of the wide gamut display to be used by multiplying by , and then compared with the result converted to the HSV color space. In the example of Figure 9(C), it can be seen that for the red hue, a positive difference in lightness V corresponds to a red color indication value R being displayed stronger than specified, and for the cyan hue, a positive difference in saturation S corresponds to a red color indication value R being displayed negative. In reality, since it is impossible to display a negative R, the color mapping described above is used to eliminate the requirement for color indication values ​​(R, G, B) in a non-standard RGB color space. Even in such cases, the auxiliary display can be used to confirm the state of the color indication. This display can be similarly performed for other color spaces that can be considered cylindrical.

[0069] The calculation of the above auxiliary displays is performed by the auxiliary display calculation program 402 of the auxiliary display calculation unit 400 in Figure 2. By looking at these auxiliary displays, the user can quantitatively determine which hues have narrower gradations than those of standard color vision, or which hues appear dark or dull to those of standard color vision. This knowledge may be useful in subsequent creative activities, social experiences, and so on. The amount of change in color indicator values ​​is obtained by taking the difference between the converted color indicator values ​​and the color indicator values ​​before and after the change, and this can be easily displayed precisely because the core of this invention is the conversion of color indicator values.

[0070] Other Embodiments The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit and scope of the present invention. For example, the following modifications are also included. (1) Various standards are known for color profiles or color spaces, and the above-described embodiment is merely an example, and the present invention is not limited to this. (2) The various devices shown in the above embodiments are merely examples and can be applied to various devices. For example, the display may be a display device of a smartphone or a tablet. (3) In the above embodiment, "image" includes various still images and moving images such as paintings, animations, photographs, and movies. [Industrial Applicability]

[0071] According to the present invention, video signals are corrected so that the cone sensitivity distributions match between major and minority color vision persons, which allows for good recognition by minority color vision persons and color vision correction that accommodates individual differences in color vision, making the present invention suitable for display devices that have inputs for various RGB channels used by minority color vision persons. In particular, since the present invention is believed to provide a better match between color perception and that of majority color vision persons than conventional methods, it is suitable for use in art appreciation using XR (cross reality) devices and LCD pen tablets. [Explanation of symbols]

[0072] 100: Image before conversion 120: Normal display device 122:Video memory 130: Display for multicolor vision sufferers 150: Color vision correction device (color correction device) 152:Video memory 160: Correction calculation unit 162: Correction conversion calculation program 164: Adjustment calculation program 200: PC 202: Camera 204: Scanner 170: Display 172:VR equipment 174,300: Tablet 310: Display 400: Auxiliary display calculation section 402: Auxiliary display calculation program

Claims

1. A color correction method for obtaining image data of a corrected image to be displayed to a minority color vision person by correcting image data of an image before correction, the color correction method comprising: correcting the image data of the image before correction so that the cone stimulation values ​​when the majority color vision person views the image before correction match the cone stimulation values ​​when the minority color vision person views the image after correction, thereby obtaining the image data of the corrected image.

2. 2. A color correction method according to claim 1, wherein the correction of said image data is calculated based on the cone sensitivity distribution of a color minority vision person and the spectral luminance distribution of said image data on an image display means.

3. 2. The color correction method according to claim 1, wherein the correction is performed after the video data before correction is converted into PCS in CMS.

4. 4. The color correction method according to claim 3, wherein the correction conversion is performed after the color indication values ​​are brought into a range that can be expressed by the display device by color mapping in the PCS before correction.

5. 2. A color correction method according to claim 1, further comprising the step of comparing the color gamut used by the video data before correction with the color gamut that can be represented by the display device in the same color space, thereby auxiliary displaying the color reproduction range that can obtain the same cone stimulus values ​​as those of a majority of color vision persons.

6. A color correction method, characterized in that the color space according to claim 5 is used to auxiliary display the color reproduction rate of a minority color vision person who can obtain the same cone stimulation values ​​as a majority color vision person.

7. A color correction device that performs color correction based on the color correction method according to any one of claims 1 to 6, a color correction means for correcting image data of the image before correction so that cone stimulation values ​​when the majority color vision person views the image before correction coincide with cone stimulation values ​​when the minority color vision person views the image after correction, thereby obtaining image data of the image after correction.

8. 8. The color correction device according to claim 7, further comprising auxiliary display means for comparing the color reproduction range for a polychromat and the color reproduction range for a minority color on the same color space and displaying the comparison result.

9. 8. The color correction device according to claim 7, wherein said color correction means comprises an adjustment means for adjusting the image data of said corrected image so that it falls within a display range of a display device.

Images