A device and method for enhancing color contrast based on human color vision characteristics

Abstract

This invention relates to user-specific color contrast enhancement. The invention proposes an apparatus and corresponding method for estimating a user's categorical color matching function (CMF) type and / or adjusting the device's color imaging workflow. Estimating the user's categorical CMF type includes: displaying a binary color test set on a screen; calculating a score for each categorical CMF type in a predetermined set of categorical CMF types based on received input information; and selecting a categorical CMF type for the user corresponding to the maximum or minimum score. Furthermore, this invention proposes an apparatus and corresponding method for creating a binary color test set.

Description

Technical Field

[0001] This invention relates to user-specific color contrast enhancement. The invention proposes an apparatus and corresponding method for estimating a user's categorical color matching function type and / or adjusting the device's color imaging workflow. Furthermore, the invention proposes an apparatus and corresponding method for creating a binary color test set. Background Technology

[0002] Color contrast is one of the key characteristics of an image, affecting observer perception and the subjective perceived quality of the image. The observer metamerism effect describes the fact that different observers may perceive colors differently under the same lighting conditions. Currently, color imaging workflows in devices are typically based on a predefined single-standard observer color matching function (CMF), which assumes all observers have the same color vision characteristics. However, each person's color vision characteristics and optimal CMF are not entirely the same.

[0003] In the past, these differences were considered negligible and therefore not considered when optimizing devices. However, modern devices have increasingly wider color gamuts. Therefore, ignoring the individual characteristics of human color vision can lead to changes or distortions in color contrast and a decrease in perceived image quality.

[0004] To achieve a wide color gamut, a very narrow spectral power density (SPD) function, also known as "display primaries," must be implemented on the display. However, because each person's CMF (Color Matrix Factor) varies individually, a narrower display SPD function may lead to changes or distortion in color contrast due to more pronounced metamerism failure. For example, color contrast may be reduced.

[0005] To reduce metamerism failure, traditional personalized color imaging processes involve grouping observers into categorical observer types and using pseudoisochromatic images to determine the categorical observer type for the user. However, pseudoisochromatic images have stringent requirements, requiring display on a special screen with a large number of display primary colors (at least 9 to 13 theoretical primary colors). Furthermore, they can only distinguish a limited number of categorical observer types.

[0006] Other conventional processes include controlling the image display device to ensure that different observers have the same perception of color and color contrast, or using other special devices, such as color blindness testing glasses.

[0007] However, in general, traditional processes require highly specialized hardware to reduce metamerism failure. Summary of the Invention

[0008] In view of the foregoing, the object of the present invention is to provide an apparatus and a corresponding method that can efficiently estimate a user's classification CMF type while having low and / or flexible hardware requirements. Furthermore, the object of the present invention is to improve the quality of perceived images and reduce the impact of color contrast distortion caused by observer metamerism failure.

[0009] These and other objectives are achieved by the inventive solutions described in the independent claims. Advantageous implementations are further defined in the dependent claims.

[0010] A first aspect of the present invention provides an apparatus for estimating a user's categorical color matching function (CMF) type. The apparatus is configured to: display a screen comprising a set of two or more binary color tests, wherein each binary color test corresponds to a pair of categorical CMF types A and B in a predetermined set of categorical CMF types, each binary color test comprising: a first image of categorical CMF type A in the pair, wherein the first image includes color contrast greater than a first threshold for user A corresponding to categorical CMF type A, and less than a second threshold for user B corresponding to categorical CMF type B; a second image of categorical CMF type B in the pair, wherein the second image includes color contrast greater than the first threshold for user B, and less than the second threshold for user A, the second threshold being a discrimination threshold, and the first threshold being equal to or greater than the second threshold; the apparatus is further configured to: calculate a score for each categorical CMF type in the predetermined set of categorical CMF types based on input information received in response to displaying the set of binary color tests on the screen; and select a categorical CMF type for the user corresponding to the maximum or minimum score.

[0011] The device provided in the first aspect can efficiently estimate a user's classification CMF type by displaying a binary color test and analyzing the corresponding user input information. Therefore, it is advantageous that the device provided in the first aspect has lower hardware and / or display requirements. For example, only one display screen is needed to determine the user's classification observer type. For example, the device can be a smartphone and / or can include only three display primary colors. The device may not require special equipment and / or other equipment; the display screen can be a smartphone or gadget screen. Compared to conventional solutions, the requirement for the number of different display primary colors on the display screen may be lower. By configuring the color imaging process according to the selected CMF type, the device can also reduce the impact of color contrast distortion caused by observer metamerism failure.

[0012] In addition, compared with traditional schemes, it can distinguish more classification CMF types and / or can more flexibly adjust the number of classification CMF types considered.

[0013] Furthermore, the device is not limited to a single binary color test set; it can also use different binary color test sets to flexibly estimate the user's CMF type.

[0014] The first threshold can be greater than the second threshold. Alternatively, the first threshold and the second threshold can be equal. For example, the first threshold and the second threshold can be the same discrimination threshold.

[0015] The first threshold and / or the second threshold can be measured in deltaE or dE according to CIEDE2000.

[0016] The discrimination threshold can be a threshold for color contrast, used to distinguish between color contrast that is perceptible to the user and that is not. For example, according to the CIEDE2000 index, the discrimination threshold can be equal to 2dE, while according to the CIEDE2000 index, the color contrast of a user in the CMF type A category can be less than 2dE, and therefore cannot be distinguished.

[0017] The discrimination threshold can be equal to the minimum color contrast between two colors that a user can distinguish and / or differentiate.

[0018] The discrimination threshold can be equal to the minimum color contrast that a user can perceive.

[0019] The first and second thresholds can be based on experimental results, for example, by testing and comparing various first and second thresholds.

[0020] In one implementation of the first aspect, for each binary color test, the input information includes one of the following: the first image is selected; the second image is selected; neither image is selected.

[0021] In another implementation of the first aspect, each score is calculated based on a weighted sum, where each addend in the weighted sum corresponds to a binary color test. If an image corresponding to the CMF type of the weighted sum is selected, the addend is equal to the weight; otherwise, the addend is equal to zero.

[0022] In another implementation of the first aspect, the predetermined set of classified CMF types is based on two or more display primary colors of the display screen, and / or in the predetermined set of classified CMF types, each different classified CMF type is paired once with each of the other classified CMF types to form a classified CMF type pair.

[0023] Alternatively, the predetermined set of CMF (Color, Material, Finish) types can be based on three or more display primary colors of the display. For example, since smartphone displays typically include exactly three display primary colors, the predetermined set of CMF types can be based on the three display primary colors of the display.

[0024] In another implementation of the first aspect, for each image: the color contrast is maximized or nearly maximized for users of the corresponding CMF category, and the color contrast is minimized or nearly minimized for users of the other CMF category in the pair.

[0025] In another implementation of the first aspect, each image includes a pair of two colors; for each category CMF type pair, the binary color test set includes M different binary color tests and 2*M different color pairs, where M is a positive integer greater than 1.

[0026] In another implementation of the first aspect, the device is used to cause the display screen to display each binary color test in the set two or more times.

[0027] A binary color test set can include each different binary color test two or more times.

[0028] In another implementation of the first aspect, each score is calculated based on a weighted sum, and the device is configured to determine the weight of each weighted sum based on one or more of the following weight subtypes: prior weight; individual posterior weight; general posterior weight; statistical posterior weight.

[0029] In another implementation of the first aspect, the device is further configured to perform one or more of the following operations: when the display screen displays each binary color test two or more times, determining an individual posterior weight based on the reliability of the user's selection of each binary color test; and / or when the display screen displays each binary color test two or more times, determining a general posterior weight based on the overall reliability of the user's selection of all binary color tests.

[0030] General posterior weights can be parameters used to estimate the reliability of a user as a responder. In the following example, each binary color test is shown to the user twice, and the general posterior weights are normalized. If the general posterior weight is closer to 1, the user's first and second answers to the same test are likely to be consistent. Therefore, the user can be considered reliable, and their answers provide reliable information. If the general posterior weight is less than 0.5, it indicates that the user's answers may be unreliable, and the color imaging workflow should be based on a standard observer.

[0031] In another implementation of the first aspect, the prior weight is determined by comparing the calculated color contrast of each image with the calculated color contrast corresponding to the corresponding same CMF type pair and the CMF type, and minimizing the objective function for the corresponding image; and / or the statistical posterior weight is determined based on statistical information on the reliability of multiple users' selection of each binary color test when the display screen displays each binary color test two or more times.

[0032] In another implementation of the first aspect, the weight of each weighted sum is equal to one of the following weights: the individual posterior weight; the individual posterior weight multiplied by the T-norm of the prior weight and the general posterior weight; the individual posterior weight multiplied by the statistical posterior weight and then multiplied by the T-norm of the prior weight and the general posterior weight.

[0033] In another implementation of the first aspect, the device is further configured to: display images based on a selected classification CMF type and images based on a standard observer CMF type; and obtain input information about which CMF type was selected.

[0034] The display will also show other binary color tests to verify the estimated CMF type for the user. If the verification fails, the user can choose the standard observer's CMF instead of the estimated CMF type, or re-estimate the user's CMF type.

[0035] In another implementation of the first aspect, the device is further configured to configure a color imaging process according to the selected CMF type.

[0036] A second aspect of the invention provides an apparatus for creating a binary color test set. The apparatus is used to: determine a categorical color matching function (color) based on two or more primary colors of a display screen. A set of matching function (CMF) types is created; one or more categorical CMF type pairs are created, wherein each categorical CMF type pair includes categorical CMF type A and categorical CMF type B, and each categorical CMF type is paired once with each other categorical CMF type in the set of categorical CMF types; a first color pair and a second color pair are determined for each categorical CMF type pair, wherein the first color pair has a color contrast greater than a first threshold for user A corresponding to categorical CMF type A, and less than a second threshold for user B corresponding to categorical CMF type B; the second color pair has a color contrast greater than the first threshold for user B, and less than the second threshold for user A, wherein the second threshold is a discrimination threshold, and the first threshold is equal to or greater than the second threshold; two images, including a first image and a second image, are created for each categorical CMF type pair, wherein the first image corresponds to categorical CMF type A and includes a pattern of the first color pair, and the second image corresponds to categorical CMF type B and includes a pattern of the second color pair; a first binary color test set is created, wherein each binary color test corresponds to a categorical CMF type pair and includes the corresponding two created images.

[0037] A binary color test set may include or consist of a first binary color test set.

[0038] The set of CMF types can be a predefined set of CMF types provided by the first aspect.

[0039] Within a set of categorical CMF types, some categorical CMF types may not be paired with every other categorical CMF type to create one or more categorical CMF type pairs. For example, the number of binary color tests in a binary color test set can be reduced by decreasing the number of categorical CMF type pairs. A smaller binary color test set may require less user input, but may result in less accurate estimations of the user's categorical CMF type.

[0040] In one implementation of the second aspect, for each image: the color contrast is maximized or nearly maximized for the user of the corresponding CMF category, and the objective function is minimized or nearly minimized by considering the color contrast of the users of the two CMF categories in the pair, and the color contrast is minimized or nearly minimized for the user of the other CMF category in the pair.

[0041] In another implementation of the second aspect, minimizing the objective function means that: the color contrast of the two CMF types of users in the pair produces at least the Nth minimum objective function output, where N is a positive integer between 2 and 10; or the color contrast of the two CMF types of users in the pair produces an objective function output less than a predetermined threshold.

[0042] Alternatively, N can be a positive integer greater than 1.

[0043] Furthermore, minimizing the objective function can be achieved by finding the local minimum of the objective function output through the color contrast of the two CMF types in the pair and / or by finding the N”'th local minimum of the objective function output through the color contrast, where N”' is a positive integer greater than 1.

[0044] Color contrast may be related to the two CMF types in the pair. For example, color contrast could refer to: the first color contrast for users of CMF type A and the second color contrast for users of CMF type B.

[0045] In another implementation of the second aspect, the device is further configured to create one or more other binary color tests for each class CMF type pair; each image of each other binary color test includes at least one different color compared to each image corresponding to the corresponding same class CMF type pair; the other binary color tests and the first binary color test set constitute a second binary color test set.

[0046] A binary color test set may include a second binary color test set or consist of it.

[0047] In another implementation of the second aspect, the device is further configured to create one or more copies of each binary color test in the first binary color test set or the second binary color test set.

[0048] A binary color test set may include all binary color tests in a first binary color test set and copies of each binary color test in the first binary color test set, or consist of them. Alternatively, a binary color test set may include all binary color tests in a second binary color test set and copies of each binary color test in the second binary color test set, or consist of them.

[0049] In another implementation of the second aspect, the device is further configured to: create prior weights for each image by: calculating a first color contrast of the corresponding color pair; calculating a second color contrast of the color pair corresponding to the corresponding same CMF type pair and the color pair corresponding to the CMF type, and minimizing the objective function; comparing the first contrast and the second color contrast; and / or, when the set includes each binary color test twice or more, creating statistical posterior weights based on statistical information on the reliability of multiple users' selections of each binary color test.

[0050] A third aspect of the invention provides a method for operating a device for estimating a user's categorical color matching function (CMF) type. The method includes: causing a display screen to display a set comprising two or more binary color tests, wherein each binary color test corresponds to a pair of categorical CMF types A and B in a predetermined set of categorical CMF types, each binary color test comprising: a first image of categorical CMF type A in the pair, wherein the first image includes color contrast greater than a first threshold for user A corresponding to categorical CMF type A, and less than a second threshold for user B corresponding to categorical CMF type B; a second image of categorical CMF type B in the pair, wherein the second image includes color contrast greater than the first threshold for user B, and less than the second threshold for user A, the second threshold being a discrimination threshold, and the first threshold being equal to or greater than the second threshold; the method further includes: calculating a score for each categorical CMF type in the predetermined set of categorical CMF types based on input information received in response to causing the display screen to display the set of binary color tests; and selecting a categorical CMF type for the user corresponding to the maximum or minimum score.

[0051] The method provided in the third aspect can be implemented in a manner corresponding to the device provided in the first aspect. The method provided in the third aspect and its implementation achieves the same advantages and effects as the device provided in the first aspect and its corresponding implementation.

[0052] A fourth aspect of the present invention provides a method for operating an apparatus for creating a binary color test set. The method includes: determining a categorical color matching function based on two or more primary colors of the display screen. A set of CMF (Color, Color, Function) types is created; one or more categorical CMF type pairs are created, wherein each categorical CMF type pair includes categorical CMF type A and categorical CMF type B, and each categorical CMF type is paired once with each other categorical CMF type in the set of categorical CMF types; a first color pair and a second color pair are determined for each categorical CMF type pair, wherein the first color pair has a color contrast greater than a first threshold for user A corresponding to categorical CMF type A, and less than a second threshold for user B corresponding to categorical CMF type B; the second color pair has a color contrast greater than the first threshold for user B, and less than the second threshold for user A, wherein the second threshold is a discrimination threshold, and the first threshold is equal to or greater than the second threshold; two images, including a first image and a second image, are created for each categorical CMF type pair, wherein the first image corresponds to categorical CMF type A and includes a pattern of the first color pair, and the second image corresponds to categorical CMF type B and includes a pattern of the second color pair; a first binary color test set is created, wherein each binary color test corresponds to a categorical CMF type pair and includes the corresponding two created images.

[0053] The implementation of the method provided in the fourth aspect can correspond to the implementation of the device provided in the second aspect. The method provided in the fourth aspect and its implementation achieves the same advantages and effects as the device provided in the second aspect and its corresponding implementation.

[0054] A fifth aspect of the present invention provides a computer program product including program code. The computer program product is used to control a device provided by the first aspect or the second aspect or any implementation thereof, or, when the program code is executed on a computer, the computer program product is used to perform a method provided by the third aspect or the fourth aspect or any implementation thereof.

[0055] Each observer may have a set of Color Filters (CMFs), which determines personalized color perception. Multiple CMFs from different observers can be grouped into categories, or categorical CMF types, where each group can be described using a CMF set. Each categorical observer type can represent a group or cluster of individual observers with similar CMFs. Each observer may correspond to only one categorical observer type, determined by each observer's individual CMF set. A user's categorical CMF type can be determined by the similarity between the user's individual CMF set and the generalized categorical CMF set. Specific signal processing procedures can be employed for each type of observer.

[0056] In this invention, the phrases “classification observer type” and “classification CMF type” can be used interchangeably.

[0057] Furthermore, in this invention, the terms "user" and "observer" can be used interchangeably.

[0058] Furthermore, in this invention, the terms "picture" and "image" can be used interchangeably.

[0059] If two binary color tests include identical images, they can be considered repeatable, identical, and / or indistinguishable. Alternatively, if two binary color tests include images with identical color pairs but different patterns, they can be considered repeatable, identical, and / or indistinguishable. Alternatively, if two binary color tests include images with only slightly different color pairs, they can be considered repeatable, identical, and / or indistinguishable.

[0060] In this invention, the term "reliable" can refer to something that is consistent with and / or matches the underlying facts or the final result. For example, a subweight may be reliable if its value is consistent with the user's classification CMF type and / or the user's final estimated classification CMF type.

[0061] It should be noted that all devices, elements, units, and components described in this invention can be implemented in software or hardware elements or any combination thereof. All steps performed by the various entities described in this invention, and the functions described as being performed by the various entities, are intended to indicate that the respective entities are suitable for or used to perform the corresponding steps and functions. Although the specific functions or steps performed by external entities are not reflected in the detailed description of the specific elements of the entities performing the specific steps or functions in the following detailed description of specific embodiments, those skilled in the art will understand that these methods and functions can be implemented by the corresponding software or hardware elements or any combination thereof. Attached Figure Description

[0062] The following description of embodiments, taken in conjunction with the accompanying drawings, will illustrate the above aspects and their implementation.

[0063] Figure 1 An embodiment of the present invention provides a device for estimating the classification CMF type of a user.

[0064] Figure 2 An apparatus for creating binary color test combinations is shown according to an embodiment of the present invention.

[0065] Figure 3 Two CMF sets are shown based on two popular standard observer definitions.

[0066] Figure 4 This invention illustrates a process for determining color pairs for observer classification, provided by one embodiment of the invention.

[0067] Figure 5 A summary of the computational steps provided in one embodiment of the present invention is shown to maximize the color contrast of a type A classification observer and minimize the color contrast of a type B classification observer.

[0068] Figure 6 The present invention illustrates the steps included in estimating the categorical observer type of a user according to an embodiment of the invention.

[0069] Figure 7 This invention illustrates a method for a color imaging workflow for personalizing color imaging and / or adjusting a device, according to one embodiment of the present invention.

[0070] Figure 8 The present invention provides a method according to one embodiment.

[0071] Figure 9 The present invention provides a method according to one embodiment. Detailed Implementation

[0072] Figure 1 A device 100 is shown for estimating the classification CMF type 101 of user 300. The device 100 is used to cause a display screen 102 to display a set 103 including two or more binary color tests 104. The device 100 may include or be connected to the display screen 102. The device 100 may be a smartphone including the display screen 102.

[0073] Each binary color test 104 includes a first image 107a of categorized CMF type A 101a and a second image 107b of categorized CMF type B 101b. CMF type A 101a and CMF type B 101b are included in a pair of 105 categorized CMF types corresponding to binary color test 104. Furthermore, CMF type A 101a and CMF type B 101b are included in a predetermined set of categorized CMF types 106.

[0074] Figure 1 The device 100 is also shown to receive input information 109 from user 300 in response to displaying a set 103 of binary color tests 104 on display screen 102. The device 100 is used to calculate a score 108 for each category CMF type 101 in a predetermined set 106 of category CMF types based on the received input information 109. Furthermore, the device is used to select for user 300 the category CMF type 101 corresponding to the maximum or minimum score 108.

[0075] The first image 107a includes color contrast, which is greater than a first threshold for user A corresponding to category CMF type A 101a, and less than a second threshold for user B corresponding to category CMF type B 101b. The second image 107b includes color contrast, which is greater than the first threshold for user B, and less than the second threshold for user A. The second threshold is a discrimination threshold, and the first threshold is equal to or greater than the second threshold.

[0076] Figure 2 A device 200 for creating a binary color test set 104 103 is shown. Device 200 may be a computing device. Device 200 may be integrated with device 100, but device 100 and device 200 are typically independent devices.

[0077] The device 200 is used to determine a set 106 of categorized CMF types 101 based on two or more display primary colors of the display screen 102, and to create one or more categorized CMF type pairs 105. Each categorized CMF type pair 105 includes categorized CMF type A 101a and categorized CMF type B 101b, and in the set 106 of categorized CMF types, each categorized CMF type 101 is paired once with each of the other categorized CMF types 101.

[0078] Furthermore, device 200 is used to determine a first color pair 201a and a second color pair 201b for each category CMF type pair 105, wherein the first color pair 201a has a color contrast greater than a first threshold for user A corresponding to category CMF type A 101a, and less than a second threshold for user B corresponding to category CMF type B 101b; the second color pair 201b has a color contrast greater than the first threshold for user B, and less than the second threshold for user A. The second threshold is a discrimination threshold, and the first threshold is equal to or greater than the second threshold.

[0079] Furthermore, device 200 is used to create two images, including a first image 107a and a second image 107b, for each CMF type pair 105, wherein the first image 107a corresponds to CMF type A 101a and includes a pattern of a first color pair 201a, and the second image 107b corresponds to CMF type B 101b and includes a pattern of a second color pair 201b. The first image 107a and the second image 107b are included in a binary color test 104.

[0080] In addition, the device 200 is used to create a first binary color test set 103a, wherein the first set 103a includes a binary color test for each category CMF type pair 105, the binary color test corresponding to the category CMF type pair 105 and including two corresponding created images 107a, 107b.

[0081] Binary color test set 103 may include or consist of a first binary color test set 103a. Binary color test set 103 may include or consist of a second binary color test set 103b. Second binary color test set 103b may include the first binary color test set 103a.

[0082] Device 200 can be used to provide the created first binary color test set 103a to device 100. In addition, device 200 can be used to provide the created second binary color test set 103b and / or the created binary color test set 103 to device 100.

[0083] The color rendering matrix (CMF) set of a specific user 300 of device 100 may differ from the CMF set of a standard observer. Therefore, optimizing the image processing workflow of device 100 for a standard observer may result in a decrease in color contrast perceived by user 300. By adjusting the color imaging workflow based on the classification CMF type 101 rather than the CMF set of a standard observer, device 100 can improve and personalize the color contrast for each specific user 300. Device 100 can estimate the classification CMF type 101 and provide a corresponding image processing workflow.

[0084] Device 100 can estimate the classification CMF type 101 for each user 300 based on the binary color test set 103 generated by device 200. Device 100 can display each binary color test 104 to the user 300 on a display screen 102, for example, on a smartphone display screen. Furthermore, device 100 can automatically estimate the classification CMF type 101 of the user 300 using classification rules for processing user responses. Some advantages provided by the present invention (especially those provided by device 100 and device 200) are described in detail below.

[0085] In modern colorimetric theory, test color stimuli can be matched using an additive mixture of three independent primary color stimuli, where no single primary color stimulus can be matched using an additive mixture of the other two primary color stimuli. A set including the three CIE standard CMFs can be used to calculate the tri-stimuli values ​​XYZ according to the following equation:

[0086]

[0087] Where P(λ) is the spectral power distribution of the emission source, It is the CIE standard color matching function (CMF), N” is the normalization factor, and I(λ) is the reference white light source.

[0088] The typical color imaging workflow of traditional equipment is based on a single standard observer and assumes that all observers have the same set of CMFs. Figure 3 Two CMF sets are shown based on the following two popular standard observer definitions: having a second field of view CIE standard colorimetric observer and with a 10-degree field of view The CIE standard colorimetric observer.

[0089] CMF type 101 can be classified according to metamerism. Observers will perceive color differently based on metamerism depending on the type of CMF classification.

[0090] A classifier can categorize an observer by finding a color pair that produces a large visual difference only for one type of classifier 101, and a small visual difference for other types of classifier 101.

[0091] Figure 4 A process 400 is illustrated for determining color pairs 201a, 201b for observer classification based on metamerism, wherein multiple color pairs including color α and color β are tested to determine color pairs 201a, 201b. Color pairs 201a, 201b optimize observer classification within a pair of classification observer types 105, including classification observer type A 101a and classification observer type B 101b. Process 400 can be executed by device 200. The input values ​​received in the first step 401 of each loop of this process are the RGB coordinates of color α and color β, and two sets of CMFs (CMFs) for the two classification CMF types 101 (type A and type B). A and CMF B ) and the display primary color (SPD) of the display screen 102.

[0092] The CIEDE2000 index can be used for visual difference estimation. Color coordinates can be defined in the sRGB space. For example, the coordinates of color α can be (R² - R² - R²)² - R ... α G α B α The coordinates of color β can be (R... β G β B β According to the SPD function (i.e., primary color) of display screen 102, SPD... R (λ), SPD G (λ), SPD B (λ), the spectrum of these colors when displayed on display screen 102 can be determined as follows:

[0093] Sp_color α (λ)=R α SPD R (λ)+G α SPD G (λ)+R α SPD R (λ)

[0094] Sp_color β (λ)=R β SPD R (λ)+G β SPD G (λ)+B β SPD B (λ)

[0095] The perceptual color difference calculated using the CIEDE2000 formula can be based on the trichromatic stimulus values ​​within the CIE 1931 XYZ color space in color spectral representation. The trichromatic stimulus values ​​for any spectral color can be obtained based on the observer's CMF (Color, Color, and Function). The XYZ coordinates of these colors can be obtained from the set of color matching functions for categorized observer type k = 101. Based on these coordinates, the CIEDE2000 color difference between the two colors can be calculated.

[0096] The set of categorized CMF types 106 can include 10 categorized CMF types 101. However, more or fewer categorized CMF types 101 can be used. On the one hand, the number of binary color tests depends on the number of categorized CMF types 101. Therefore, it is best to reduce the number of categorized CMF types 101. On the other hand, increasing the number of categorized CMF types 101 can achieve more accurate classification.

[0097] The CIEDE2000 index can be calculated using two colors, α and β, which represent the visual color contrast of a classification observer type k = 101.

[0098]

[0099] Among them, x_cat k (λ), y_cat k (λ), z_cat k (λ) is the CMF of a classifier of type k 101.

[0100] Figure 4 The second step 402 is shown, which includes calculating the distance between color α and color β of classifying observer type A101a and classifying observer type B101b according to CIEDE2000. and

[0101] Step 401 and step 402 may be repeated once or multiple times to calculate the color distance between multiple color pairs, including color α and color β, respectively.

[0102] Figure 4 The third step 403 includes determining D A It is the maximum value, D B Let α and β be the two colors that minimize the value. This means performing an optimization task on all possible pairs of colors α and β.

[0103] To find two colors α and β such that the classification observer type A 101a To achieve the maximum value, the function for solving this minimization task can be defined as follows:

[0104]

[0105] Furthermore, for colors α and β, the pairs 201a and 201b (the initial coordinates of color α are (R...) α G α B α The initial coordinates of color β are (R... β G β B β The CIEDE2000 index of the XYZ coordinates obtained through the CMF of other categorical observer types 101 (e.g., categorical observer type B 101b) should be minimized or nearly minimized.

[0106] and

[0107] Where, ΔE thr It is a threshold.

[0108] Based on these conditions, an objective function can be created that maximizes the visual color contrast of one type of classifier 101 (e.g., type A 101a) and minimizes the visual color contrast of other type of classifier 101 (e.g., type B 101b).

[0109] Optimization tasks may include finding colors α and β that minimize or nearly minimize the objective function. For example, for observer pair 105 of classifying observer type A101a (cat_A) and classifying observer type B101b (cat_B), the objective function can be defined according to the following equation, where cat_A is the target classifying observer:

[0110]

[0111] F(x)=(x-thr max )2

[0112] Threshold thr max This can be determined experimentally. Based on the Huawei Mate 20 Pro SPD, it targets three threshold values ​​(thr). max Experiments (i.e., 1, 2, and 5) solving the above optimization problem show that the color contrast between the two colors for the target observer (observer A) and the non-target observer (observer B) may not change significantly. Therefore, using a very large threshold may not be recommended. Based on further analysis of the obtained test images, a feasible threshold, thr, was determined. max =2. Other thresholds can also be used.

[0113] Determining the color pairs 201a and 201b can be achieved by minimizing or nearly minimizing the following objective function:

[0114]

[0115] By minimizing or nearly minimizing the objective function, two colors 201a can be identified. These two colors have high color contrast for observers classifying CMF type A101a, but low color contrast for observers classifying CMF type B101b, i.e., they are similar in color.

[0116] Minimizing the objective function can refer to determining the variable parameters of the objective function, such as color pairs and classification CMF type pairs 105, or the color contrast of user 300 of classification CMF type A 101a and the color contrast of user 300 of classification CMF type B 101b, thereby producing one or more of the following: at least the Nth smallest objective function output; objective function output less than a predetermined threshold; a local minimum of the objective function output; the Nth"'th smallest local minimum of the objective function output.

[0117] N can be a positive integer greater than 1, N”' can be a positive integer greater than 1, and the predetermined threshold can be determined by testing and / or comparing multiple threshold iterations and / or by experiment.

[0118] The variable parameters that minimize or nearly minimize the objective function (e.g., color pairs and classification CMF type pairs 105, or color contrast) can be determined through testing and / or comparison of multiple variable parameters iteratively and / or experimentally. For example, multiple objective function outputs for multiple color contrasts can be computed, and these multiple objective function outputs can be compared with each other and / or compared with one or more of the above requirements.

[0119] Figure 4 The fourth step 404 involves drawing a special pattern (i.e., a picture) using the color pair 201a obtained in the third step 403, which can distinguish between observer type A 101a and observer type B 101b. Process 400 can be repeated to create a pattern using other color pairs 201b, which makes the distance D... B Maximize and make distance D A Minimize. Therefore, two images 107a and 107b can be created for each CMF type pair 105. In addition, the creation of two images 107a and 107b as described above can be repeated for other classifier observer type pairs 105, wherein each pair of images 107a and 107b can form a binary color test 104.

[0120] The minimization task can be solved for all possible pairs of categorical observer types, N". The number of pairs of categorical observer types N can be determined by the following combinatorial equation:

[0121]

[0122] Where K is the number of classifier observer types 101. Therefore, the total number of different test images in the binary color test sets 103 and 103a can be 2N, etc. Each classifier CMF type pair 105 corresponds to a binary color test 104, and each binary color test 104 corresponds to two images 107a and 107b.

[0123] Binary color test set 103 may include multiple different binary color tests 104 for each category observer type pair 105. Furthermore, binary color test set 103 may also include copies and / or slightly modified copies of each different binary color test 104.

[0124] Binary color tests 104 can be created based on defined color pairs 201a and 201b. Each binary color test 104 may include two circular images 107a and 107b, where each circle may include a symbol or pattern on a background. In the first circle 107a, the color difference between the symbol and the background may be obvious to an observer of classification type A 101a, but not obvious to an observer of classification type B 101b. In the second image 107b, the color difference between the symbol and the background may be obvious to an observer of classification type B 101b, but not obvious to an observer of classification type A 101b. Various different shapes and patterns can be used.

[0125] Figure 5 A summary of the computational steps described above is shown to find a color pair 201a that maximizes the color contrast of a classification observer of type A 101a and minimizes the color contrast of a classification observer of type B 101b. The computational steps can be performed by device 200.

[0126] An observer can categorize images based on user input information 109 associated with binary color test set 103. Observer categorization can be performed by device 100. The categorization rules can be based on pairwise categorization. Based on the categorization results of all image pairs 107a and 107b (i.e., binary color test 104), it can be determined whether user 300 belongs to one of the categories, i.e., classification CMF type 101.

[0127] Experimental results show that the classification rule for determining observer type 101 is best implemented based on the "vote count" for each observer type 101. The "vote count" refers to the number of times user 300 selects the image corresponding to a specific observer type 101 with more significant color contrast during the test.

[0128] Each binary color test 104 can be displayed to the user 300 times. Alternatively, for each binary color test 104, multiple binary color tests 104 can be created, including pairs of the same color 201a and 201b but with different patterns.

[0129] For each binary color test 104, user 300 can select the first image 107a, the second image 107b, or neither image, for example, skipping binary color test 104. Matching parameters can be introduced based on one or more selections by user 300. In a preferred example, each different binary color test 104 can be displayed to the user 300 twice. Therefore, the user input information 109 for each different binary color test 104 can include two answers: answer1 and answer2, that is, each identical binary color test 104 corresponds to one answer.

[0130] Matching parameters It can be defined as follows:

[0131]

[0132] Suppose that the same binary color test 104 is shown to user 300 twice, and each time the same image 107a corresponding to categorical observer type A101a is selected instead of image 107b corresponding to categorical observer type B101b. Then the matching parameters of the two binary color tests 104 are... Both can be 0.5. When calculating score 108 and estimating the classification CMF type 101 for user 300, two binary color tests 104 can be considered. Therefore, the sum of the matching parameters of the two binary color tests 104 could be 1, representing the ideal reliability of user 300 for binary color test i 104.

[0133] Furthermore, the score 108 corresponding to category observer type A 101a will increase, while the score 108 corresponding to category observer type B 101b will remain unchanged. Assuming user 300 selects image 107b corresponding to category observer type B 101b, the score 108 corresponding to category observer type B 101b will increase, while the score 108 corresponding to category observer type A 101a will remain unchanged.

[0134] Generally speaking, the matching parameters for each binary color test 104 are the same as those for binary color test 104i. It can be defined as follows:

[0135]

[0136] Wherein, N1 is the number of times user 300 selects image 107a corresponding to category observer type A 101a when displaying binary color test i 104 or one of the same binary color test i 104; N2 is the number of times user 300 selects image 107b corresponding to category observer type B 101b when displaying binary color test i 104 or one of the same binary color test i 104; and N3 is the number of times binary color test i 104 or the same binary color test i 104 is displayed on display screen 102 and / or shown to user 300.

[0137] The matching parameters can be normalized based on the number of copies of each different binary color test 104 in the binary color test set 103.

[0138] The normalization method for matching parameters may differ from the example above.

[0139] The matching parameters for each identical binary color test 104 can be the same.

[0140] For example, the matching parameters can be defined in other ways, where a reliable choice increases the score by 108 compared to an unreliable choice.

[0141] For example, the matching parameters can be defined in other ways, where a reliable choice would reduce the score by 108 compared to an unreliable choice.

[0142] Matching parameters It can also be called individual posterior weights.

[0143] The score of 108 for the categorical observer type A101a(Cat_type[A]) can be calculated based on the number of votes as follows:

[0144]

[0145] R_weight_i can be based on one or more other sub-weights corresponding to the binary color test i 104 and the same binary color test 104 as the binary color test i 104.

[0146] Alternatively, the score 108 for Cat_type[A] can be defined as follows:

[0147]

[0148] Among them, all sub-weight types are merged into the total weight corresponding to binary color test i 104 and binary color test 104 which is the same as binary color test i 104.

[0149] The total weight may depend on the user's 300 responses to each of the 104 binary color tests. For each of the 104 binary color tests, there can be two total weights: and At least one total weight in the total weight of each binary color test 104 can represent minimum reliability, for example, it can be equal to 0. Therefore, one total weight in the total weight of each binary color test 104 can be ignored and / or not added to the score 108, thus not increasing the score 108.

[0150] Sub-weights may depend on user 300's answers to each binary color test 104. For each binary color test 104, each sub-weight type can have two sub-weights, for example, and And / or and

[0151] The weights for each binary color test 104 can be based on prior subweights and posterior subweights.

[0152] Prior weights can be based on estimates based on the CIEDE2000 index.

[0153] Other binary color tests 104 can be used and / or created; for example, other binary color tests 104 can be added to the first binary color test set 103a. The other binary color tests 104 may include images based on other color pairs that correspond to local minima of the objective function and / or nearly minimize the objective function.

[0154] For each image, a color pair and its CIEDE2000 index can be used. To determine the prior weights, the CIEDE2000 index corresponds to the image's corresponding CMF type 101, and the image's corresponding CMF type 101 and CMF type pair 105 correspond to the global minimum of the objective function.

[0155] Furthermore, other color pairs and their CIEDE2000 indices can be used to determine prior weights. These other color pairs may correspond to local minima of the objective function and / or may minimize the objective function based on the corresponding classification CMF type 101 and classification CMF type pair 105 of the image. In one example, the other color pairs corresponding to the second and third local minima of the objective function can be determined based on the CIEDE2000 indices. Therefore, the total number of different binary color tests 104 in the binary color test sets 103 and 103a can be increased for statistical verification. Other binary color tests and the first binary color test set 103a can constitute the second binary color test set 103b.

[0156] Compared to the pair corresponding to the global minimum of the objective function, the second and third color pairs may have lower reliability and / or a smaller effect on estimating the CMF type 101 for user 300. This can be taken into account by introducing corresponding prior weights. The prior weights can be calculated as follows:

[0157]

[0158] Posterior weights can include one or more of the following sub-weights: individual posterior weights, general posterior weights, and statistical posterior weights.

[0159] Generally, the posterior weights can be based on the reliability of the user input information 109. Information regarding the reliability of the binary color test set 103 for a specific user 300 can be obtained from the experimental results of the specific user 300 performing the binary color test set 103. In a preferred example, each binary color test 104 can be displayed twice. Matching parameters can be used based on the user 300's response.

[0160]

[0161] Generally, the posterior weights can be determined as follows:

[0162]

[0163] Where N' is the total number of different binary color tests 104 in the binary color test set 103. The general posterior weight can be the average of the individual posterior weights. Therefore, the general posterior weight can represent the general reliability of a specific user 300 and the corresponding user input information 109, based on all binary color tests 104 in the binary color test set 103. For example, if user 300 randomly selects an image from each binary color test 104, according to the general posterior weight, the user's overall reliability is likely to be considered low. In this case, the user input information 109 may not provide accurate information about user 300's classification CMF type 101 and may therefore be ignored. For a specific user 300, the general posterior weight of all binary color tests 104 in the binary color test set 103 may be the same.

[0164] The statistical posterior weights can be based on the reliability of user input information 109 from multiple users 300 for each binary color test 104. Experiments may show that some binary color tests 104 are statistically less reliable than others. Binary color tests 104 with lower reliability can be replaced by binary color tests 104 with higher reliability. Furthermore, the statistical posterior weights of binary color tests 104 with lower reliability may be smaller. Therefore, the impact of binary color tests 104 with lower reliability on estimating the classification CMF type 101 may be reduced.

[0165] The statistical posterior weights can be determined as follows:

[0166]

[0167] Where N' is the total number of different binary color tests 104 in the binary color test set 103, and M' is the total number of users 300 considered in the statistical evaluation of the binary color test set 103.

[0168] One or more sub-weights can be combined into a corresponding total weight. For example, each sub-weight can be multiplied to determine each total weight. Alternatively, one or more sub-weight types may not be used.

[0169] Sub-weights can be merged with one or more other weight subtypes based on the T-norm in fuzzy logic theory:

[0170]

[0171] Sub-weights can be merged with one or more other weight subtypes based on other functions, for example,

[0172]

[0173] For example, the total weight can be defined as follows:

[0174]

[0175] Individual posterior weights and general posterior weights can be determined by device 100, for example, in a smartphone. Prior weights and statistical posterior weights can be determined by device 200, for example, in an external processing device. This external processing device has access to more processing power and / or more data than device 100. Weights calculated in device 200 can be provided to device 100. Alternatively, all weights can be determined in device 100.

[0176] Figure 6The steps that may be included when estimating the classification observer type 101 for user 300 are illustrated. These steps can be performed by device 100. In this example, the binary color test set 103 is based on five classification observer types 101. Each classification observer type 101 is paired once with each of the other classification observer types 101, resulting in 10 classification observer type pairs 105. The binary color test set 103 is displayed to user 300, wherein each different binary color test 104 is displayed twice in a random order. Therefore, a total of 20 binary color tests 104 are displayed to user 300. Each binary color test 104 includes two images 107a and 107b, thus providing a selection opportunity for each binary color test 104.

[0177] also, Figure 6 An exemplary binary color test i 104 is shown, comprising two circles, each circle containing a pattern. It is worth noting that the binary color test 104 is sensitive to color. Figure 6 The exemplary images shown are for reference only. Device 100 may display a binary color test 104 on display screen 102. User 300 may choose whether to prefer the right circle or the left circle, or whether the binary color test 104 should be skipped. If user 300 perceives a higher color contrast in one image than another and / or perceives a clearer pattern in one image than another, the first image may be preferred. For example, the pattern could be numbers or letters. If user 300 cannot determine which image is preferred, the binary color test 104 may be skipped. Figure 6 In the exemplary binary color test 104 shown, the left image 107a corresponds to the category observer type A101a, and the right image 107b corresponds to the category observer type B101b.

[0178] Based on user 300's answer, device 100 receives user input information 109 (response). Based on the user input information 109, weights can be calculated for the binary color test 104. Matching parameters can be calculated based on two separate answers received from user 300 regarding the exemplary binary color test 104. For example, if user 300 prefers the left image 107a in both answers, then the matching parameters... It can be as large as and / or equal to 0.5, while the matching parameter It can be very small and / or equal to 0. Furthermore, other sub-weights can be calculated. By multiplying the matching parameter by these other sub-weights and then adding the result to the score 108, the score 108 for classifying observer type A 101a can be increased. On the other hand, due to the matching parameter... The value can be equal to 0, so the score 108 of the categorical observer type B 101b may not increase. The above-mentioned adjusted score 108 can be repeated for each binary color test 104 in the binary color test set 103. Based on the final score 108, user 300 can be assigned to the category with the maximum value of the corresponding score 108 (i.e., the weighted sum).

[0179] According to one embodiment of the present invention, a method may include a first step and a second step. For example, the first step may be performed during smartphone (device 100) initialization or during other calibration processes, including displaying a binary color test set 103 and acquiring user input information 109; the second step may be applied to the color imaging process of device 100, wherein the standard observer CMF may be replaced by a determined classification CMF type 101. The method may be performed by device 100.

[0180] A set of binary color tests 103 can be displayed to the user of device 100, wherein each binary color test 104 can consist of two images 107a and 107b. Furthermore, each image 107a and 107b may include symbols on a background, wherein each image 107a and 107b can be determined according to two classification CMF types 101a and 101b. The number of binary tests 104 can be determined based on the severity of metamerism failure on the display screen 102 displaying the binary color tests 104.

[0181] User 300 can select images 107a and 107b where the perceived color difference between the background and the symbol is more pronounced. For example, if user 300 does not perceive a color contrast advantage in one image 107a or 107b relative to the other, user 300 can refuse to select it. User 300 can visually compare the distinguishability of the stimulus to the background in the two images 107a and 107b of each binary color test 104.

[0182] Based on the automatic analysis of user 300's answers to the binary color test set 103, user 300's classification CMF type 101 can be estimated, wherein the classification CMF type 101 most similar to user 300's individual CMF type 101 can be selected. Furthermore, determining user 300's classification CMF type 101 can be based on one or more sub-weights. These sub-weights can represent the reliability of the binary color test 104 and the reliability of user 300's answers.

[0183] Based on the identified CMF type 101, a special test image can be generated and displayed to the user 300 to verify the test results.

[0184] Based on the estimated classification CMF type 101 for user 300, a new color imaging workflow can be implemented to take into account personalized color reproduction, wherein the color workflow can be adjusted from the color workflow corresponding to the standard CMF type observer to the color workflow corresponding to the estimated classification CMF type 101.

[0185] Figure 7 A method 700 for a color imaging workflow for personalizing color imaging and / or adjusting a device is illustrated. Method 700 can be performed by device 100. Imaging input data for visualizing an image on display 102 of device 100 can be acquired 701 using a general device-independent format (e.g., sRGB format). The imaging input data can undergo a standard correction process 702, including one or more of the following processes: linearization (gamma correction), reference white point (D65) consideration, white balance, color transformation, and / or other processes, and can be converted 703 to XYZ format.

[0186] Colors can be rendered on display screen 102 using the primary colors of display screen 102. The final spectrum emitted by display screen 102 can be perceived by user 300, where user 300's perception depends on user 300's individual CMF. For example, CMF variations among people are caused by the individual distribution of receptors on the retina. Furthermore, based on user 300's input information 109, user 300's categorical observer type can be estimated.

[0187] For personalized color correction 704, the imaging data can be converted from the XYZ format used for a standard observer to the XYZ' format used for a specific observer type 705, where the personal characteristics of user 300 can be taken into account. Furthermore, the imaging data can be converted to a display color space 706. The final data can then undergo a standard correction process again, and can be rendered 707 and finally displayed on the display screen 102.

[0188] Figure 8 A method 800 according to an embodiment of the present invention is illustrated. Method 800 can be performed by device 100. Method 800 includes step 801: causing a display screen to show a set comprising two or more binary color tests. Furthermore, method 800 includes step 802: calculating a score for each category CMF type in a predetermined set of category CMF types based on input information received in response to causing the display screen to show the binary color test set. Furthermore, method 800 includes step 803: selecting a category CMF type for user 300 corresponding to the maximum or minimum score.

[0189] Figure 9A method 900 according to an embodiment of the present invention is illustrated. Method 900 can be performed by device 200. Method 900 includes step 901: determining a set of categorized CMF types based on two or more display primary colors of the display screen. Furthermore, method 900 includes step 902: creating one or more pairs of categorized CMF types, wherein each pair of categorized CMF types includes categorized CMF type A and categorized CMF type B. Furthermore, method 900 includes step 903: determining a first color pair for each pair of categorized CMF types, wherein the first color pair has a color contrast greater than a first threshold for user A corresponding to categorized CMF type A, and less than a second threshold for user B corresponding to categorized CMF type B. Furthermore, method 900 includes step 904: determining a second color pair for each pair of categorized CMF types, wherein the second color pair has a color contrast greater than the first threshold for user B, and less than the second threshold for user A. Furthermore, method 900 includes step 905: creating two images for each CMF type pair, including a first image and a second image, wherein the first image corresponds to CMF type A and includes a pattern of a first color pair, and the second image corresponds to CMF type B and includes a pattern of a second color pair. Furthermore, method 900 includes step 906: creating a first binary color test set, wherein each binary color test corresponds to a CMF type pair and includes the corresponding two created images.

[0190] Experiments show that 88% of users selected the CMF type, which was individually determined according to an embodiment of the present invention, as the preferred CMF type. That is, approximately 88% of users indicated that color contrast was improved after color transformation based on the individually determined CMF type.

[0191] After color correction for the corresponding classifiers, the CIEDE2000 color difference increased by an average of 1.75 times, enabling users to distinguish colors that were indistinguishable before the color correction process.

[0192] According to one embodiment of the present invention, the classification CMF type can be determined based on color pairs with small color differences in CIEDE2000. Therefore, compared with conventional methods, the requirement for the number of different display primary colors of the display screen can be reduced and / or decreased.

[0193] This invention has been described in conjunction with various embodiments as examples and implementations. However, based on a study of the drawings, the invention, and the independent claims, those skilled in the art will be able to understand and implement other variations when practicing the claimed subject matter. In the claims and the description, the word "comprising" does not exclude other elements or steps, and the quantifier "a" does not exclude a plurality. A single element or other unit may fulfill the function of several entities or items described in the claims. Listing certain measures in dissimilar dependent claims does not imply that combinations of these measures cannot be used in advantageous implementations.

Claims

1. A device (100) for estimating the classification color matching function (CMF) type (101) of a user (300), characterized in that, The device (100) is used for: The display screen (102) displays a set (103) of two or more binary color tests (104). Each binary color test (104) corresponds to a pair (105) of categorical CMF type A (101a) and categorical CMF type B (101b) in a predetermined set (106) of categorical CMF types, and each binary color test (104) includes: The first image (107a) of the CMF type A (101a) in the pair (105) includes a color contrast greater than a first threshold for user A corresponding to CMF type A (101a), and less than a second threshold for user B corresponding to CMF type B (101b). The second image (107b) of the classification CMF type B (101b) in the pair (105), wherein the second image (107b) includes color contrast, the color contrast being greater than the first threshold for user B and less than the second threshold for user A, the second threshold being a discrimination threshold, and the first threshold being equal to or greater than the second threshold; The device (100) is also used for: Based on the input information (109) received in response to displaying the binary color test (104) set (103) on the display screen (102), a score (108) is calculated for each of the CMF types (101) in the predetermined set (106) of the CMF types. Select the category CMF type (101) corresponding to the maximum or minimum score (108) for the user (300).

2. The device (100) according to claim 1, characterized in that, For each binary color test (104), the input information (109) includes one of the following: the first image (107a) is selected; the second image (107b) is selected; or neither image is selected.

3. The device (100) according to claim 2, characterized in that, Each score (108) is calculated based on a weighted sum, where each addend in the weighted sum corresponds to a binary color test (104). If the image corresponding to the classification CMF type (101) of the weighted sum is selected, the addend is equal to the weight; otherwise, the addend is equal to zero.

4. The device (100) according to claim 1, characterized in that, The predetermined set (106) of the classification CMF types is based on two or more display primary colors of the display screen (102), and / or in the predetermined set (106) of the classification CMF types, each different classification CMF type (101) is paired once with each other classification CMF type (101) to form a classification CMF type pair.

5. The device (100) according to claim 1, characterized in that, For each image: The color contrast is maximized or nearly maximized for users of the corresponding CMF type (101) category. The color contrast is minimized or nearly minimized for users of another CMF type (101) in the pair.

6. The device (100) according to claim 1, characterized in that, Each image includes a pair of two colors; for each CMF type pair, the binary color test set (103) includes M different binary color tests (104) and 2*M different color pairs, where M is a positive integer greater than 1.

7. The device (100) according to claim 1, characterized in that, The device (100) is used to cause the display screen (102) to display each binary color test (104) in the set (103) two or more times.

8. The device (100) according to claim 1, characterized in that, Each score (108) is calculated based on a weighted sum, and the device (100) is used to determine the weight of each weighted sum according to one or more of the following weight subtypes: Prior weights; Individual posterior weights; General posterior weights; Statistical posterior weights.

9. The device (100) according to claim 7 or 8, characterized in that, The device (100) is also used to perform one or more of the following operations: When the display screen (102) displays each binary color test (104) two or more times, individual posterior weights are determined based on the reliability of the user's (300) selection of each binary color test (104); and / or When the display screen (102) displays each binary color test (104) two or more times, a general posterior weight is determined based on the overall reliability of the user's (300) selection of all binary color tests (104).

10. The device (100) according to claim 8, characterized in that, The prior weights are determined for the corresponding images by comparing the calculated color contrast of each image with the calculated color contrast corresponding to the same CMF type pair (105) and the corresponding CMF type (101), and minimizing the objective function; and / or The statistical posterior weight is determined based on statistical information about the reliability of multiple users' selection of each binary color test (104) when the display screen (102) displays each binary color test (104) two or more times.

11. The device (100) according to claim 8, characterized in that, The weight of each weighted sum is equal to one of the following weights: The individual posterior weights; The individual posterior weight is multiplied by the T-norm of the prior weight and the general posterior weight; The individual posterior weight is multiplied by the statistical posterior weight, then multiplied by the T-norm of the prior weight and the general posterior weight.

12. The device (100) according to claim 1, characterized in that, The device (100) is also used for: The display screen (102) displays images based on the selected classification CMF type (101) and images based on the standard observer CMF type; Obtain input information (109) about which CMF type (101) was selected.

13. The device (100) according to claim 1, characterized in that, The device (100) is also used to configure the color imaging process according to the selected CMF type (101).

14. An apparatus (200) for creating a set (103) of binary color tests (104), characterized in that, The device (200) is used for: The set of color matching function (CMF) types (101) (106) is determined based on two or more primary colors of the display screen (102). Create one or more categorized CMF type pairs (105), wherein each categorized CMF type pair (105) includes categorized CMF type A (101a) and categorized CMF type B (101b), and in the set of categorized CMF types (106), each categorized CMF type (101) is paired once with each other categorized CMF type (101); For each CMF type pair (105), a first color pair (201a) and a second color pair (201b) are determined, wherein the first color pair (201a) has a color contrast greater than a first threshold for user A corresponding to the CMF type A (101a) and less than a second threshold for user B corresponding to the CMF type B (101b), and the second color pair (201b) has a color contrast greater than the first threshold for user B and less than the second threshold for user A, wherein the second threshold is a discrimination threshold and the first threshold is equal to or greater than the second threshold; For each CMF type pair (105), create two images including a first image (107a) and a second image (107b), wherein the first image (107a) corresponds to the CMF type A (101a) and includes the pattern of the first color pair (201a), and the second image (107b) corresponds to the CMF type B (101b) and includes the pattern of the second color pair (201b); Create a first set of binary color tests (104) (103a), wherein each binary color test (104) corresponds to a classified CMF type pair (105) and includes the two corresponding created images.

15. The device (200) according to claim 14, characterized in that, For each image: The color contrast is maximized or nearly maximized for users of the corresponding CMF type (101) category. The objective function is minimized or nearly minimized by considering the color contrast of users of two different CMF types in the CMF type pair, wherein the color contrast is minimized or nearly minimized for users of the other CMF type (101) in the CMF type pair.

16. The device (200) according to claim 15, characterized in that, Minimizing the objective function means: The color contrast of the two CMF types in the pair produces at least the Nth minimum objective function output, where N is a positive integer between 2 and 10; or The color contrast of the two CMF types of users in the pair produces an objective function output that is less than a predetermined threshold.

17. The device (200) according to any one of claims 14 to 16, characterized in that, The device (200) is also used to create one or more other binary color tests (104) for each CMF type pair. Compared to each image corresponding to the corresponding CMF type pair of the same classification, each image of each other binary color test (104) includes at least one different color; The other binary color tests (104) and the first binary color test set (103a) constitute the second binary color test set (103b).

18. The device (200) according to claim 17, characterized in that, The device (200) is also used to create one or more copies of each binary color test (104) in the first binary color test set (103a) or the second binary color test set (103b).

19. The device (200) according to claim 15, characterized in that, The device (200) is also used for: Create prior weights for each image using the following steps: Calculate the first color contrast of the corresponding color pair; Calculate the second color contrast of the color pairs corresponding to the corresponding same CMF type pairs (105) and CMF type (101), and minimize the objective function; and Compare the first color contrast and the second color contrast; and / or When the set (103) includes each binary color test (104) twice or more, statistical posterior weights are created based on statistical information on the reliability of the selection of each binary color test (104) by multiple users.

20. A method (800) for operating a device (100) for estimating a user's classification color matching function (CMF) type (101), characterized in that, The method (800) includes: Make the display (801) (102) display a set (103) of two or more binary color tests (104). Each binary color test (104) corresponds to a pair (105) of categorical CMF type A (101a) and categorical CMF type B (101b) in a predetermined set (106) of categorical CMF types, and each binary color test (104) includes: The first image (107a) of the CMF type A (101a) in the pair includes a color contrast that is greater than a first threshold for user A corresponding to CMF type A (101a), and less than a second threshold for user B corresponding to CMF type B (101b). The second image (107b) of the classification CMF type B (101b) in the pair, wherein the second image (107b) includes color contrast, the color contrast being greater than the first threshold for user B and less than the second threshold for user A, the second threshold being a discrimination threshold, and the first threshold being equal to or greater than the second threshold; The method (800) further includes: Based on the input information (109) received in response to displaying the binary color test set (103) on the display screen (102), a score (802) is calculated (108) for each of the CMF types (101) in the predetermined set (106) of the CMF types. Select (803) the classification CMF type (101) corresponding to the maximum or minimum score for the user.

21. A method (900) for operating an apparatus (200) for creating a set (103) of binary color tests (104), characterized in that, The method (900) includes: The set of color matching function (CMF) types (101) is determined (901) based on two or more primary colors of the display screen (102); Create (902) one or more class CMF type pairs, wherein each class CMF type pair (105) includes class CMF type A (101a) and class CMF type B (101b), wherein in the set of class CMF types, each class CMF type (101) is paired once with each other class CMF type (101); For each CMF type pair, a first color pair and a second color pair (903, 904) are determined, wherein the first color pair has a color contrast greater than a first threshold for user A corresponding to the CMF type A (101a) and less than a second threshold for user B corresponding to the CMF type B (101b), and the second color pair has a color contrast greater than the first threshold for user B and less than the second threshold for user A, wherein the second threshold is a discrimination threshold and the first threshold is equal to or greater than the second threshold; For each CMF type pair, create (905) two images including a first image (107a) and a second image (107b), wherein the first image (107a) corresponds to the CMF type A (101a) and includes a pattern of the first color pair, and the second image (107b) corresponds to the CMF type B (101b) and includes a pattern of the second color pair; Create (906) a first binary color test set, wherein each binary color test (104) corresponds to a classified CMF type pair (105) and includes the two corresponding created images.

22. A computer program product including program code, characterized in that, The computer program product is used to control the device according to any one of claims 1 to 19, or when the program code is executed on a computer, the computer program product is used to perform the method (800, 900) according to claim 20 or 21.

Images