A method, device and product for determining skin color and adjusting ink amount of printed characters

Abstract

The application discloses a printing character favorite skin color determination and ink amount adjustment method, device and product, relates to the printing technical field, and obtains different visual effect skin color images as the images to be evaluated by limiting the maximum ink amount and adjusting the ink amount combination of each channel; favorite probability of each image to be evaluated is acquired to determine the favorite skin color of the printing character and K ink reduction amount; the favorite color image chroma value is used to draw a chroma scatter diagram to fit the favorite ellipse of different color deviations; the C ink reduction amount, M ink reduction amount and Y ink reduction amount are determined based on the favorite ellipse, and the printing of the favorite skin color image is realized according to the determined favorite skin color of the printing character and the ink reduction amount. The application can quickly and accurately determine the favorite skin color and reduce the ink cost of the printing enterprise.

Description

Technical Field

[0001] This invention relates to the field of printing technology, and in particular to a method, apparatus and product for determining the preferred skin tone (skin color) of printed figures and adjusting the amount of ink. Background Technology

[0002] Memory color refers to people's deep memory of the colors of familiar objects (such as skin tone, sky color, fruit color, etc.); when evaluating printed images, people also focus on the color reproduction effect of these areas. In the process of color reproduction of printed images, people's evaluation of memory color is an important standard for measuring the accuracy of color reproduction. However, the preferred memory color of a familiar object in people's minds is not exactly the same as the actual surface color of that object, and may even differ significantly. Therefore, adjusting the color of the memory object to within the user's preferred color range can significantly improve the color quality of printed images.

[0003] In studies on skin tone preferences, the unevenness of skin tone makes it difficult to accurately obtain skin tone chromaticity information. Patent CN104720813A discloses a method for obtaining a skin tone color chart, using a spectrophotometer to measure the chromaticity value of a smooth area of ​​the real cheek to represent skin tone. Jiangtao Kuang, in his literature, uses the color of the forehead to represent the overall skin tone. Due to the unevenness of skin tone, using the color of one area to represent the entire skin tone may introduce some error. Patent CN113259658A discloses a method for obtaining a preferred color memory of an object. This method involves rendering a set of memory color photographs and organizing observers to conduct psychophysical experiments to select preferred colors, measuring the average spectral reflectance at different locations in the skin photographs to represent the reflectance of the real skin. Patent CN101911118A discloses a skin tone evaluation method that divides the entire face into at least 25 regions of different colors, and represents the chromaticity value of each region by calculating the average chromaticity value of pixels within each region. In 2022, Krishnapriya determined skin regions by setting a threshold and calculating the average chromaticity value of pixels within the region as skin color. In 2023, Peng Rui obtained skin color chromaticity values ​​by calculating the average of all pixel colors when characterizing displayed skin color.

[0004] In actual printing, the above methods are difficult to quickly and accurately reproduce preferred skin tones. Patents with publication numbers CN104720813A and CN113259658A establish a color-matching chart based on visual evaluation experiments, allowing image colors to be adjusted to closely match the chart's colors in practical applications. However, since the color-matching chart only provides a reference for preferred skin tones, printing companies cannot achieve real-time adjustments from the output skin tone image to the desired skin tone during the printing process. Considering the production costs and efficiency of printing companies, the timeliness of image skin tone adjustment methods is somewhat limited. Summary of the Invention

[0005] The purpose of this invention is to provide a method, apparatus, and product for determining preferred skin tone and adjusting ink volume in printed portraits, so as to quickly and accurately determine preferred skin tone and reduce ink costs for printing companies.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] A method for determining preferred skin tone and adjusting ink volume in printed portraits, the method comprising:

[0008] Acquire digital images containing information about the person; the digital images include S skin tone images of different ages and shooting environments; the printing color mode of the skin tone images is CMYK; C represents cyan; M represents magenta; Y represents yellow; K represents black;

[0009] For any skin tone image in the digital image, an H-width ink volume limiting image is generated; the maximum ink volume of each H-width ink volume limiting image is different.

[0010] For any ink volume limiting image, the ink volume combinations of the single-channel, dual-channel, and triple-channel ink volume limiting image are adjusted respectively to generate N1+N2+N3 ink volume adjustment images; N1 represents the number of single-channel ink volume adjustment images; N2 represents the number of dual-channel ink volume adjustment images; N3 represents the number of triple-channel ink volume adjustment images.

[0011] Each ink volume limitation image is used as an image to be evaluated, and each ink volume adjustment image is used as an image to be evaluated, generating S×H×(N1+N2+N3+1) images to be evaluated.

[0012] For any of the images to be evaluated, the face of the image to be evaluated is segmented to obtain skin color regions, and the skin colors in the skin color regions are clustered to obtain clustering results;

[0013] For any of the images to be evaluated, feature color patches of the image to be evaluated are constructed based on the clustering results, and the average chromaticity value of all feature color patches is calculated to obtain the chromaticity value of the image to be evaluated; in the clustering results, one skin color corresponds to one feature color patch;

[0014] Obtain the preference probability of each of the images to be evaluated, and calculate the average preference probability of the ink volume adjustment images with the same maximum ink volume, so as to obtain the average preference probability of each ink volume limit image.

[0015] The N1+N2+N3 ink volume adjustment images generated from the ink volume limitation image with the highest average preference probability are used as the target image set, and the skin color in the target image set is used as the preferred skin color for printed figures.

[0016] Based on the ink volume of the target image, adjust the ink volume combination and preference probability of the image, and determine the amount of ink reduction (K).

[0017] The images to be evaluated with the highest probability of preference, ranked T, are used as preferred color images, and a chromaticity scatter plot is drawn based on the chromaticity values ​​of the preferred color images; T represents the average number of images preferred by each observer when they judge whether they like the images to be evaluated.

[0018] Based on the chromaticity scatter plot, fit the preferred ellipse for different color biases, and reduce the major and minor axes of the preferred ellipse to obtain the preferred reduced ellipse.

[0019] Obtain the preference level of each point on the preference ellipse and the reduced preference ellipse, and determine the chromaticity value of each sample on the preference ellipse and the reduced preference ellipse; the sample is the point on the chromaticity scatter plot corresponding to the center of the preference ellipse, the endpoints of the major and minor axes, and the endpoints of the major and minor axes of the reduced preference ellipse.

[0020] Based on the preference level and the chromaticity values ​​of each sample, the C ink reduction, M ink reduction, and Y ink reduction of the target image set ink volume adjustment image are determined; the preferred skin tone of the printed person, the K ink reduction, the C ink reduction, the M ink reduction, and the Y ink reduction are used to print the preferred skin tone image.

[0021] Optionally, for any ink volume limiting image, the ink volume combinations of the single-channel, dual-channel, and three-channel ink volume limiting image are adjusted respectively to generate N1+N2+N3 ink volume adjustment images, specifically including:

[0022] For any ink-limited image, the ink volume of each single channel of the ink-limited image is reduced by a different proportion to obtain N1 single-channel ink-adjustment images; the single channel is one of the C channel, M channel, Y channel and K channel;

[0023] For any ink-limited image, the ink volume is reduced by a different proportion in each dual channel of the ink-limited image to obtain N2 dual-channel ink-adjusted images; the dual channels are two of the C channel, M channel, Y channel and K channel;

[0024] For any ink-limited image, reduce the ink volume of the K channel of the ink-limited image by a fixed proportion, and at the same time reduce the ink volume of both the C channel, M channel and Y channel of the ink-limited image by different proportions, to obtain N3 three-channel ink-limited adjustment images.

[0025] Optionally, the skin colors in the skin-colored regions are clustered to obtain clustering results, specifically including:

[0026] An improved K-means algorithm is used to calculate the weighted Euclidean distance from each pixel in the skin color region to the cluster center; the improved K-means algorithm is a K-means algorithm with added weights; the weights are determined based on the contour coefficients of the pixels in the skin color region;

[0027] The weighted Euclidean distance is used to determine the skin color category of each pixel in the skin color region, and the clustering result is obtained.

[0028] Optionally, the ink amount combination and preference probability of the image are adjusted according to the ink amount in the target image set to determine the K ink reduction amount, specifically including:

[0029] Compare the difference in preference probabilities between the ink amount adjustment images with reduced ink amount (K inks) and the ink amount adjustment images without reduced ink amount (K inks) in the target image set.

[0030] If the difference is within the set difference range, then K ink needs to be reduced, and the value of the amount of K ink reduction needs to be determined.

[0031] If the difference exceeds the set difference range, then there is no need to reduce K ink, and the amount of K ink reduction is zero.

[0032] Optionally, based on the chromaticity scatter plot, a preference ellipse for different color biases is fitted, and the major and minor axes of the preference ellipse are reduced to obtain a reduced preference ellipse, specifically including:

[0033] Based on the chromaticity scatter plot, fit the whiter preference ellipse, the yellower preference ellipse, and the redder preference ellipse.

[0034] Reduce the major and minor axes of the whitish preference ellipse to half their original size to obtain a whitish preference minimized ellipse. Reduce the major and minor axes of the yellowish preference ellipse to half their original size to obtain a yellowish preference minimized ellipse. Reduce the major and minor axes of the reddish preference ellipse to half their original size to obtain a reddish preference minimized ellipse.

[0035] Optionally, the formula for calculating the weighted Euclidean distance is:

[0036]

[0037] Where, d 1,2 This represents the weighted Euclidean distance from pixel 'a' in the skin color region to the cluster center; This represents the brightness of the color of pixel 'a' in the skin color region. This represents the color saturation of pixel 'a' in the skin color region; The hue of the color of pixel 'a' in the skin color region; The brightness of the color representing the cluster centers; The saturation of the color representing the cluster centers; The hue represents the color of the cluster centers; m is the weight of the saturation difference; n is the weight of the hue difference. and It is calculated based on the red-green and blue-yellow hues of pixel a in the skin color region; and It is calculated based on the red-green and blue-yellow hues of the cluster centers.

[0038] Optionally, before calculating the average chromaticity values ​​of all feature color patches to obtain the chromaticity values ​​of the image to be evaluated, the method further includes:

[0039] The chromaticity values ​​of each characteristic color patch were measured using a spectrophotometer with a geometric measurement condition of 45 / 0; the chromaticity values ​​included at least: red-green chromaticity and blue-yellow chromaticity.

[0040] Optionally, S≧4; H≧2; the maximum ink consumption range is [260%, 340%].

[0041] The present invention also provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for determining the preferred skin tone and adjusting the ink volume of printed figures.

[0042] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for determining the preferred skin tone and adjusting the ink volume of printed figures.

[0043] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0044] This invention, through limiting the maximum ink usage and adjusting the ink volume combination of each channel, obtains skin tone images with different visual effects as evaluation images. By determining the preference probability of each evaluation image, the preferred skin tone and the amount of K ink reduction for printed figures are determined. By using the chromaticity values ​​of the preferred color image to draw a chromaticity scatter plot, the preference ellipses of different color biases are fitted, thereby determining the amount of C ink reduction, M ink reduction, and Y ink reduction. This invention can quickly and accurately determine the preferred skin tone and reduce the ink costs for printing companies. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of the method for determining the preferred skin tone and adjusting the ink volume of printed figures according to Embodiment 1 of the present invention.

[0047] Figure 2 This is an illustration of skin tone in a human figure.

[0048] Figure 3 A schematic diagram of the skin color region segmentation results for S1-S5;

[0049] Figure 4 A schematic diagram showing the results of masking the non-skin-colored areas in S1-S5;

[0050] Figure 5 For the preferred color image a * 10 b * 10 Scatter plot, schematic diagram of the fitting results of three preferred ellipses: "whiter", "yellower", and "redder".

[0051] Figure 6 This is a schematic diagram showing the preference ranking results for the three preference ellipses: "whiter", "yellower", and "redder".

[0052] Figure 7 This is a diagram illustrating the method and size adjustment for converting a disliked skin tone image to a liked image.

[0053] Figure 8 This is a diagram of the internal structure of a computer device. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] The purpose of this invention is to provide a method, apparatus, and product for determining preferred skin tone and adjusting ink volume in printed portraits, aiming to quickly and accurately determine preferred skin tone and reduce ink costs for printing companies.

[0056] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] Example 1

[0058] like Figure 1 As shown, the method for determining the preferred skin tone and adjusting the ink volume of printed figures in this embodiment includes:

[0059] Step 101: Obtain digital images containing information about the person; the digital images include S skin tone images of different ages and shooting environments.

[0060] The printing color mode of the skin color image is CMYK; C represents cyan; M represents magenta; Y represents yellow; and K represents black.

[0061] Step 102: For any skin tone image in the digital image, generate an H-width ink volume limit image; the maximum ink volume of each H-width ink volume limit image is different.

[0062] Step 103: For any ink volume limiting image, adjust the ink volume combination of the single channel, dual channel and three channel of the ink volume limiting image respectively to generate N1+N2+N3 ink volume adjustment images.

[0063] Where N1 represents the number of single-channel ink volume adjustment images; N2 represents the number of dual-channel ink volume adjustment images; and N3 represents the number of three-channel ink volume adjustment images.

[0064] Step 104: Use each ink volume limitation image as an image to be evaluated, and use each ink volume adjustment image as an image to be evaluated, to generate S×H×(N1+N2+N3+1) images to be evaluated.

[0065] Step 105: For any of the images to be evaluated, the face of the image to be evaluated is segmented to obtain the skin color region, and the skin color in the skin color region is clustered to obtain the clustering result.

[0066] Step 106: For any of the images to be evaluated, construct feature color patches of the image to be evaluated based on the clustering results, and calculate the average chromaticity value of all feature color patches to obtain the chromaticity value of the image to be evaluated.

[0067] In the clustering results, one skin color corresponds to one feature color block.

[0068] Step 107: Obtain the preference probability of each of the images to be evaluated, and calculate the average preference probability of the ink volume adjustment images with the same maximum ink volume, so as to obtain the average preference probability of each ink volume-limited image.

[0069] Step 108: Take the N1+N2+N3 ink volume adjustment images generated from the ink volume limitation image with the highest average preference probability as the target image set, and take the skin color in the target image set as the preferred skin color for printing the person.

[0070] Step 109: Adjust the ink volume combination and preference probability of the image based on the ink volume of the target image set, and determine the amount of ink reduction K.

[0071] Step 110: Select the top T images with the highest liking probability as the preferred color images, and draw a chromaticity scatter plot based on the chromaticity values ​​of the preferred color images.

[0072] Where T represents the average number of images that each observer likes when judging whether they like the images being evaluated.

[0073] Step 111: Fit different color preference ellipses based on the chromaticity scatter plot, and reduce the major and minor axes of the preference ellipses to obtain a reduced preference ellipse.

[0074] Step 112: Obtain the preference level of each point on the preference ellipse and the reduced preference ellipse, and determine the chromaticity value of each sample on the preference ellipse and the reduced preference ellipse. The sample refers to the point on the chromaticity scatter plot corresponding to the center of the preference ellipse, the endpoints of its major and minor axes, and the endpoints of the major and minor axes of the reduced preference ellipse.

[0075] The preference level was determined through multiple preference ranking experiments.

[0076] Step 113: Determine the C ink reduction, M ink reduction, and Y ink reduction of the target image set ink volume adjustment image based on the preference level and the chromaticity value of each sample.

[0077] The preferred skin tone of the printed person, the reduction amount of K ink, the reduction amount of C ink, the reduction amount of M ink, and the reduction amount of Y ink are used to print the preferred skin tone image.

[0078] In one example, in step 101, digital images with minimal facial obscuration and taken from the front of the subject are selected whenever possible.

[0079] Different ages refer to three different age groups: children, youth, and the elderly; different shooting environments refer to different lighting sources, including indoor lighting sources and outdoor natural light sources at different times of day.

[0080] If the number of skin color images S ≥ 4, then an image may contain multiple different people selected as subjects of study, and in this case, an image can be split into multiple images.

[0081] In one example, step 102 specifically includes: setting color management parameters for the skin tone image in the digital image in Photoshop software, limiting the maximum ink usage of the image output, and generating H ink limit images with different maximum ink usage, that is, the maximum ink usage type is H.

[0082] In this step, the maximum ink usage (total ink volume in printing) is limited by establishing the ICC profile of the printing equipment, and the color conversion method is perceptible. The gray component substitution (GCR) method can be used to replace part of the black in the overprinting of yellow, magenta, and cyan inks with black ink, which can produce a more beneficial visual effect or a more ink-saving image without affecting the visual effect.

[0083] In order to determine a visually appealing maximum ink usage limit effect in subsequent steps, the maximum ink usage type in this example is H≧2, and the maximum ink usage value range is [260%, 340%].

[0084] In one example, step 103 specifically includes:

[0085] For any ink-limited image, the ink volume of each single channel of the ink-limited image is reduced by a different proportion to obtain N1 single-channel ink-adjusted images; the single channel is one of the C channel, M channel, Y channel and K channel.

[0086] For any ink-limited image, the ink volume of each dual channel of the ink-limited image is reduced by a different proportion to obtain N2 dual-channel ink-adjusted images; the dual channels are two of the C channel, M channel, Y channel and K channel.

[0087] For any ink-limited image, reduce the ink volume of the K channel of the ink-limited image by a fixed proportion, and at the same time reduce the ink volume of both the C channel, M channel and Y channel of the ink-limited image by different proportions, to obtain N3 three-channel ink-limited adjustment images.

[0088] In this example, the CMYK single-channel, dual-channel, and triple-channel ink volume combinations of the H-frame ink volume limitation image generated in step 102 are adjusted to generate N1 single-channel ink volume adjustment images, N2 dual-channel ink volume adjustment images, and N3 triple-channel ink volume adjustment images, forming S×H×(N1+N2+N3+1) images to be evaluated. More specifically:

[0089] In step 103, for images that have passed the total ink limit, it is necessary to reduce the ink volume of the CMYK single channel, dual channel, and triple channel respectively to obtain skin color samples with richer visual effects and further achieve the purpose of saving ink.

[0090] In step 103, the images with different maximum ink usage limits for the H-frame generated in step 102 are used as the original images. Under the condition of not exceeding the maximum ink usage setting, the ink usage of each channel is reduced. This can be done by reducing the ink usage of a single channel -C, -M, -Y, -K; reducing the ink usage of a dual channel -CM, -CY, -YM, -CK, -MK, -YK; or reducing the ink usage of a three-channel -CMK, -CYK, -YMK.

[0091] In step 103, to prevent image color distortion caused by excessive ink reduction when adjusting the ink volume of dual-channel and triple-channel, the ink volume of the K channel is fixedly reduced by 5% when combined with other channels.

[0092] In step 103, all images use a uniform representation method to indicate their corresponding maximum ink usage adjustment and CMYK channel ink reduction. For example, with a maximum ink usage limit of 280%, a 3% reduction in C channel ink, and a 5% reduction in Y channel ink, the representation method is: 280-C-3-Y-5.

[0093] In one example, step 105 specifically includes: ① segmenting the skin color region in Photoshop software, or setting up a skin color detection program in self-developed software to segment the skin color region and save the skin color region as an image; ② reading the segmented image (i.e., the skin color region) in the software and using a skin color clustering model to cluster the skin color into class X.

[0094] In step ① of this example, the purpose of segmenting the skin-colored region is to avoid non-skin-colored regions in the image, such as clothing, hair, eyes, and lips, affecting the clustering results of skin color. You can use Photoshop's selection tool to select and save the entire facial skin-colored region. Alternatively, you can use programming software such as Matlab to set the skin-colored region recognition algorithm. For example, the Adobe RGB chromaticity value range for the skin-colored region can be set to: R = 220-240, G = 180-200, B = 170-190. The skin-colored region can be saved in a digital image format such as PNG or JPG, which does not affect the reproduction of skin color.

[0095] In step ② of this example, the skin color clustering model is a model determined based on the improved K-means algorithm. Specifically, clustering the skin colors within the skin color region to obtain the clustering results includes:

[0096] An improved K-means algorithm is used to calculate the weighted Euclidean distance from each pixel in the skin color region to the cluster center; the improved K-means algorithm is a K-means algorithm with added weights; the weights are determined based on the contour coefficients of the pixels in the skin color region.

[0097] The weighted Euclidean distance is used to determine the skin color category of each pixel in the skin color region, and the clustering result is obtained.

[0098] The improved K-means algorithm will be described in more detail below.

[0099] With CIE L * C * H * This represents the color information of a pixel, where L * C represents the lightness or brightness of a color. * H represents the saturation of a color. * For the hue of a color, the formula for calculating the weighted Euclidean distance from it to the cluster center is:

[0100]

[0101] Where, d 1,2 This represents the weighted Euclidean distance from pixel 'a' in the skin color region to the cluster center; This represents the brightness of the color of pixel 'a' in the skin color region. This represents the color saturation of pixel 'a' in the skin color region; The hue of the color of pixel 'a' in the skin color region; The brightness of the color representing the cluster centers; The saturation of the color representing the cluster centers; The hue of the color representing the cluster centers; For a class of skin color pixels, the average CIE L * C * H * Chromaticity value; m is the weight of saturation difference; n is the weight of hue difference; and It is calculated based on the red-green and blue-yellow hues of pixel a in the skin color region; and It is calculated based on the red-green and blue-yellow hues of the cluster centers.

[0102] Silhouette coefficient is a clustering evaluation metric that considers both the tightness between similar clusters and the separation between different clusters. Its value ranges from -1 to 1, with values ​​closer to 1 indicating better clustering. Each pixel has a silhouette coefficient. The optimal clustering effect is achieved when the average silhouette coefficient of all pixels is maximized. The weights m and n are then determined at this optimal point. The silhouette coefficient S of pixel a... mn The calculation formula is:

[0103]

[0104] Where D in For intra-class differences, D out The difference between classes is calculated using the following formula:

[0105]

[0106]

[0107] Where p and q are the number of pixels of the same category and the closest different category to pixel a, respectively, i and j represent the i-th and j-th pixels of the same category and different categories, respectively, and d a,i d a,j This represents the distance between two pixels of the same or different categories, calculated using formula (1). By continuously updating the coordinates of the cluster centers, the clustering results are output when the coordinates of all cluster centers no longer change or when the number of iterations is reached.

[0108] In one example, before calculating the average chromaticity values ​​of all feature color patches in step 106 to obtain the chromaticity values ​​of the image to be evaluated, the method further includes: measuring the chromaticity values ​​of each feature color patch using a spectrophotometer with a geometric measurement condition of 45 / 0; the chromaticity values ​​include at least: red-green chromaticity and blue-yellow chromaticity.

[0109] Step 106 specifically includes: ① For any of the images to be evaluated, constructing feature color blocks of the image to be evaluated in the typesetting software based on the clustering results, arranging the feature color blocks near the corresponding skin tone image, and printing them out simultaneously; ② Measuring the L of each feature color block using a spectrophotometer. * a * b * The chromaticity value is calculated by taking the average of the chromaticity values ​​of all feature color patches, and then obtaining the chromaticity value of the image to be evaluated; where L * Indicates the brightness of a color, a * Indicates red-green hue, b * Indicates the degree of blue-yellow tint.

[0110] In step ① above in this example, the clustering result is X CIE Ls. * a * b* The values ​​need to be converted to RGB values. Based on the RGB values, the clustering results of an image can be represented by X uniform color blocks of 2cm×2cm in the typesetting software.

[0111] In step ② of this example, the color measuring instrument is a spectrophotometer commonly used in the printing industry with a geometric measurement condition of 0 / 45, a measurement condition of D50 / 10°, a measurement wavelength range of 400nm-700nm, and a wavelength interval not exceeding 10nm. X characteristic color patches of an image are measured, and X sets of L values ​​are recorded. * 10 a * 10 b * 10 The values ​​are averaged to represent the skin color of an image. * 10 a represents the brightness of a color when the viewing angle is 10°. * 10 b represents the red-green hue at a viewing angle of 10°. * 10 This indicates the blue-yellow tint when the viewing angle is 10°.

[0112] In one example, the preference probability in step 107 is determined through multiple preference judgment experiments. Step 107 specifically includes: ① covering non-skin-colored areas in the printed image with a black baffle to prevent observers from being influenced by the colors of other non-skin-colored areas; ② organizing observers to conduct visual experiments, where observers make preference judgments on images with different ink volume settings, with each image being judged at least D times; ③ calculating the preference probability P% for each image, where P% = (number of preference judgments / total number of observations) × 100%; ④ calculating the average preference probability P% for the same image with the maximum ink volume limit, obtaining the average preference probability for each ink volume-limited image, with the maximum ink volume limit having the highest average preference probability being suitable for printing skin-colored preferences.

[0113] In step ① above in this example, black paper is used to cover the non-skin-colored area, and the window is set according to the size of the skin-colored area.

[0114] In step ② above in this example, the observer needs to conduct the experiment under standard observation conditions, with an observation distance of 25-40cm, such as under sunlight through a north-facing window, or in a standard observation box with a color rendering index greater than 95, an illuminance of 500lx-1000lx, and a color temperature of 5000K±500K. All observers participating in the experiment have normal color vision and good color discrimination ability.

[0115] In step ② above in this example, the observer needs to conduct repeated experiments at different time periods to ensure the accuracy of the experiment; the number of preference judgments for each image is D≧30.

[0116] In step ② above in this example, the visual experiment is a preference judgment experiment. The observer can sort and compare images of the same image content printed under different ink volume conditions using a comparative method; or evaluate the preference of each printed image; the evaluation criterion is the observer's personal preference for the skin color of the people in the image; the observer judges the preference of each image and gives an evaluation of "like" or "dislike", without limiting the number of "like" or "dislike" samples.

[0117] In step ③ above in this example, a higher liking probability P% indicates that more observers like the skin color of the people under that image parameter setting.

[0118] In step ④ of this example, the average value of the probability P% of liking the same image under the maximum ink usage limit is calculated. The maximum ink usage limit method with the highest average value is suitable for the total ink usage of the printing company. Suitable for the total ink usage of the printing company refers to the total ink usage that can achieve cost savings in total ink usage without affecting the observer's perception of image liking. When the average liking probabilities of several total ink usage limits are close, the printing method that saves more ink is recommended.

[0119] In one example, step 109 specifically includes:

[0120] Compare the preference probabilities of the images with reduced ink amount (K-ink) and images with unchanged ink amount (K-ink) in the target image set. If the difference is within a set range, K-ink needs to be reduced, and the amount of reduction is determined. If the difference exceeds the set range, K-ink does not need to be reduced, and the amount of reduction is zero.

[0121] For example, to determine whether reducing K ink can increase the probability of liking or save ink without significantly reducing the probability of liking, one can compare the difference in the probability of liking between the image output without reducing K ink and the corresponding image output with reduced K ink. For example, one can compare the difference in the probability of liking between images with the following K ink values: -C, -M, -Y, -CM, -CY, -MY and images with the following K ink values: -CK, -MK, -YK, -CMK, -CYK, -MYK.

[0122] It should be noted that those skilled in the art can flexibly choose the size of the difference range, such as [0, 1%], [0, 2%], etc., which will not be elaborated here.

[0123] In one example, in step 110, the experiment is a preference judgment experiment, so there is no limit to the number of preference samples selected by the observer. However, in order to use a certain number of preference samples to fit the skin color preference ellipse in the experimental results, it is necessary to count the average number of preference samples selected by the observer.

[0124] In one example, step 111 specifically includes:

[0125] Based on the chromaticity scatter plot, fit ellipses representing white, yellow, and red preferences. Reduce the major and minor axes of the white preference ellipse to half their original values ​​to obtain a reduced white preference ellipse; reduce the major and minor axes of the yellow preference ellipse to half their original values ​​to obtain a reduced yellow preference ellipse; reduce the major and minor axes of the red preference ellipse to half their original values ​​to obtain a reduced red preference ellipse. Therefore, this example uses the plotting of T preference color images to obtain the desired ellipse. * 10 b * 10 The scatter plot was used to fit three preferred ellipses based on whether the skin tone was "lighter", "yellower", or "redder" compared to normal skin tone. By reducing the major and minor axes of the three preferred ellipses to half their original lengths, three reduced preferred ellipses were obtained.

[0126] For the above-mentioned preference ellipse fitting process, since the observer's preferred skin tone is not singular and is also influenced by factors such as age and shooting environment, it is necessary to determine three different preference ellipses—"fairer," "yellower," and "redder"—based on the visual experiment results. The ellipse fitting method is as follows:

[0127]

[0128] Its standardized ellipse equation is:

[0129]

[0130] Formula equation (6) describes a (a c b c An ellipse with center α, major axis length 2γ, and minor axis length 2β. c b c The initial values ​​of γ and β are:

[0131]

[0132]

[0133]

[0134]

[0135] Where a_data and b_data are all the preferred color samples selected in the experiment. * 10 b * 10 Colorimetric value.

[0136] The goal of ellipse fitting is to adjust a c b c The parameters γ and β minimize the residuals of all given data points with respect to the equation of this ellipse.

[0137] Let (a) i b i If a data point is a preferred image, then the residual R is... i Represented as:

[0138]

[0139] Method for calculating the sum of squared residuals:

[0140]

[0141] Where RSS is the residual sum of squares, v is the number of data points to be fitted, i is the i-th preference sample, and Ri is the sum of squares of the residuals. i Let be the residual of the i-th preferred sample. The ellipse fit is best when the sum of squared residuals is minimized.

[0142] In one example, steps 112 and 113 determine nine samples from the major and minor axes and centers of the three preference ellipses ("whiter", "yellower", and "redder") and the three preference reduction ellipses. The nine samples are sorted according to preference and classified into preference levels. The direction and size of color adjustment are determined, and the reduction amount of C, M, and Y is determined.

[0143] Since the observer's color preference distribution for different images inside the preference ellipse is not uniform, it is necessary to select several samples evenly inside the preference ellipse to classify preference levels. This classification can provide a method for adjusting the amount of ink for color adjustment of skin color samples inside the preference ellipse.

[0144] The above is based on the a of the disliked image and the like ellipse. * b * The ink volume is adjusted based on the distribution.

[0145] This embodiment addresses the uneven distribution of skin tones and the difficulty in quickly and accurately reproducing preferred skin tones. The proposed method for determining preferred skin tones and adjusting ink volume in printed portraits can directly guide ink volume adjustments during the printing process while reducing ink costs for printing companies. Specifically, it creates visually rich images by limiting the maximum ink volume of the image and reducing the ink volume of different channels. This method is simple and efficient. A skin tone clustering model is used to obtain uneven skin tone colors and establish uniform color feature blocks. This method ensures good clustering results for skin tones while simplifying the measurement process. This embodiment generates a large number of images by reducing ink volume and determines preferred colors based on visual experimental results. This improves skin tone preference while reducing ink costs for printing companies. Based on the color difference between the output printed portrait image and the chromaticity distribution differences between different levels of preferred colors within the skin tone preference ellipse, the type and magnitude of ink volume adjustment for the image color are determined. This directly guides ink volume adjustments during the printing process, improving the timeliness and accuracy of preferred skin tone reproduction.

[0146] The following is a specific example to further explain in detail the process of determining the preferred skin tone and adjusting the ink volume of the above-mentioned printing figures in practical applications.

[0147] By reducing the amount of ink used in printing to generate printed samples with different color rendering effects, a large number of observers were organized to conduct a visual evaluation experiment on preferred skin tones. Based on the experimental results, the preferred skin tone ellipse was determined, thus realizing the above method. The specific steps are as follows:

[0148] (1) Two ISO400 images and one iStock image library standard image were selected as research images. Yellow-skinned women were chosen as the research subjects in all images, resulting in five research images (S1-S5). Figure 2 The images are shown in sections (a), (b), and (c). S1, S2, and S4 represent young women, S3 represents a female child, and S5 represents an elderly woman. S1 and S2 were shot indoors, while S3, S4, and S5 were shot outdoors in natural midday light.

[0149] (2) In Photoshop, the total ink volume of digital images is limited to a maximum of 280% and 330%, and the ink volume limitation method used is gray component substitution (GCR).

[0150] (3) Using the two images with a total ink volume of 280% and 330% as the original images, reduce the ink volume of each CMYK channel.

[0151] Single channel: The ink volume of channels C, M, Y, and K is reduced by 3%, 5%, and 7% respectively. Channel K has two additional adjustment methods of 9% and 11%, forming N1=14 single channel ink volume reduction methods.

[0152] Dual-channel: C, M, and Y are reduced by 3% and 7% respectively, while K channel is reduced by a fixed 5%. C, M, Y, and K channels are combined in pairs to reduce ink volume. There is also a combination method where C channel is reduced by 9% and K channel is reduced by 5%, resulting in a total of N2=19 dual-channel ink volume reduction methods.

[0153] Three channels: C, M, and Y are reduced by 3% and 7% respectively. When C, M, and Y channels are combined in pairs, K channel is reduced by an additional 5% of ink consumption, resulting in a total of N3 = 12 ways to reduce ink consumption in three channels.

[0154] A total of 460 images with different skin tone rendering effects were generated, consisting of 5 types of images × 2 total ink volume limits × (14 single-channel + 19 dual-channel + 12 three-channel + 1 original image). S1-S5 each have 92 images with different visual effects for the observer to choose from.

[0155] (4) Use the selection tool in Photoshop to segment the skin-toned areas. The segmentation results for S1-S5 are as follows: Figure 3 As shown, Figure 3 Parts (a), (b), (c), (d), and (e) in the image correspond to the segmentation results of S1, S2, S3, S4, and S5, respectively. The skin color areas shown are all facial areas, and the skin color gamut is saved in PNG image format.

[0156] (5) Read the segmented images in Matlab software and use a skin color clustering model to cluster each skin color into 3 categories. The skin color clustering model is based on the improved K-means method, which will not be described in detail here.

[0157] (6) Based on the clustering results, convert the clustering results in step (5) into Adobe RGB values. Based on the Adobe RGB values, assign values ​​in Illustrator software to create feature color blocks. Place the feature color blocks near the right side of the corresponding output image of the person. The color blocks will be printed synchronously with the digital image of the person's skin color. Use three 2cm×2cm uniform color blocks to represent the skin color of an image.

[0158] (7) Measure the characteristic color patch L using a spectrophotometer with geometric measurement conditions of 45 / 0. * 10 a * 10 b * 10 Chromaticity values ​​are measured from three characteristic color patches of an image, and L is recorded. * 10 a * 10 b * 10 Calculate the value of L in 3 groups. *10 a * 10 b * 10 The average value. In practical applications, this example uses an X-rite Exact45 / 0 spectrophotometer to measure colorimetric values.

[0159] (8) Figure 4 As shown in sections (a), (b), and (c), black paper is used to cover non-skin-colored areas, and the window size is set according to the size of the skin-colored area in the image.

[0160] (9) Fourteen young observers were organized to participate in the visual experiment. All observers were students majoring in printing at Beijing Institute of Printing. Six of them conducted two experiments and eight conducted three experiments. The number of judgments for each image was 6×2+8×3=36.

[0161] The LED View Ultimate multi-channel standard observation box was used to simulate the D50 standard light source, with a color temperature in the range of 5000±100K, a color rendering index greater than 97, and an illuminance of about 700lx. The light source was incident perpendicularly, and the observer's gaze was directed at a 45° angle to the sample surface, with the observation point about 30cm away from the sample surface.

[0162] (10) Calculate the probability of liking each image P% = (number of times a preference was selected / total number of observations) × 100%.

[0163] (11) The average preference probability of all observers for the 230 images with a total ink volume limit of 280% was 29.7%, and the average preference probability of the 230 images with a total ink volume limit of 330% was 28.4%. The images with a total ink volume limit of 280% had a higher preference probability, and 280% is also a more ink-efficient printing method. Therefore, it is recommended to use 280% total ink volume for printing skin-tone images.

[0164] Comparing the preference probabilities of images with -C, -M, -Y, -CM, -CY, and -MY versus images with -CK, -MK, -YK, -CMK, -CYK, and -MYK, the average preference probability of images without additional reduction of K channel ink volume is 17.1%, while the average preference probability of images with reduced K channel ink volume is 16.2%. This indicates that reducing K channel ink volume has little impact on skin tone image preference. Therefore, it is recommended to reduce K ink volume by 5% for printing skin tone images.

[0165] (12) The average number of preferred images for each of S1-S5 by the observer was calculated to be 8.3. Finally, the top 8 images with the highest probability of preference were recommended as preferred color images.

[0166] (13) Draw the a for each of the eight preferred color images selected from images S1-S5. * 10 b * 10 The chromaticity scatter plot, based on the distribution of 40 chromaticity scatter points, can fit three preferred ellipses: "whiter", "yellower", and "redder". The specific representation of the ellipse fitting function will not be elaborated here.

[0167] The ellipse fitting results for the three preferences of "whiter", "yellower", and "redder" are as follows: Figure 5 As shown. Among them, the "whitish" preference oval is suitable for all ages and shooting conditions, the "reddish" preference oval is only suitable for outdoor children's pictures, and the "yellowish" preference oval is only suitable for indoor young women's pictures;

[0168] (14) Reduce the major and minor axes of the ellipse to half their original length. Nine samples are determined from the major and minor axes and the center of the three preferred ellipses: "whitish," "yellowish," and "reddish." These nine samples are then ranked according to their preference probability, and preference levels are assigned. The desired skin tone effect is determined based on age and shooting environment, and the preferred ellipse is determined accordingly. The results are shown in the appendix. Figure 6 As shown.

[0169] (15) Determine the reduction amount of C, M, and Y based on the color adjustment direction and adjustment size. Figure 7 This document describes methods for adjusting the C, M, and Y ink levels for six undesirable images. An ellipse representing the preferred color region is used, divided into nine preference levels. To bring undesirable image 1 into the preferred ellipse, the ink levels in both the M and Y channels need to be reduced by approximately 7%. To increase the preference level, the reduction in M ​​and Y channel ink levels needs to be fine-tuned to achieve the desired effect for image a. * 10 b * 10 The color coordinates are exactly at the location of the preferred ellipse point 1.

[0170] The above method obtains skin tone images with different visual effects by limiting the maximum ink usage and reducing the ink volume of CMYK subchannels. Based on visual experiment results, the observer's preferred skin tone and the ink volume adjustment method for that preferred skin tone are determined. Therefore, printed images output using this method can improve the pleasingness of skin tone images while saving ink usage for printing companies. To accurately quantify skin tone information, an improved clustering method is used to achieve the best clustering effect for color information in different locations of facial skin tone, improving the accuracy of skin color acquisition. Determining the ink volume adjustment type and size based on the difference from the preferred ellipse chromaticity distribution directly guides ink volume adjustments during the printing process; this method is simple and efficient.

[0171] Example 2

[0172] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for determining the preferred skin tone and adjusting the ink volume of a printed person in Embodiment 1.

[0173] Example 3

[0174] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for determining the preferred skin tone and adjusting the ink volume of printed figures in Embodiment 1.

[0175] Example 4

[0176] A computer program product includes a computer program that, when executed by a processor, implements the steps of the method for determining the preferred skin tone and adjusting the ink volume of a printed person in Embodiment 1.

[0177] Example 5

[0178] A computer device, the internal structure of which can be shown in the diagram below. Figure 8 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores pending transactions. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements the method for determining the skin tone preference and adjusting the ink volume of printed figures in Embodiment 1.

[0179] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this invention are all information and data authorized by the object or fully authorized by all parties.

[0180] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided by this invention may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided by this invention may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0181] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0182] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for determining preferred skin tone and adjusting ink volume in printed portraits, characterized in that, The method includes: Acquire digital images containing information about the person; the digital images include S skin tone images of different ages and shooting environments; the printing color mode of the skin tone images is CMYK; C represents cyan; M represents magenta; Y represents yellow; K represents black; For any skin tone image in the digital image, an H-width ink volume limiting image is generated; the maximum ink volume of each H-width ink volume limiting image is different. For any ink volume limiting image, the ink volume combinations of the single-channel, dual-channel, and triple-channel ink volume limiting image are adjusted respectively to generate N1+N2+N3 ink volume adjustment images; N1 represents the number of single-channel ink volume adjustment images; N2 represents the number of dual-channel ink volume adjustment images; N3 represents the number of triple-channel ink volume adjustment images. Each ink volume limitation image is used as an image to be evaluated, and each ink volume adjustment image is used as an image to be evaluated, generating S×H×(N1+N2+N3+1) images to be evaluated. For any of the images to be evaluated, the face of the image to be evaluated is segmented to obtain skin color regions, and the skin colors in the skin color regions are clustered to obtain clustering results; For any of the images to be evaluated, feature color patches of the image to be evaluated are constructed based on the clustering results, and the average chromaticity value of all feature color patches is calculated to obtain the chromaticity value of the image to be evaluated; in the clustering results, one skin color corresponds to one feature color patch; Obtain the preference probability of each of the images to be evaluated, and calculate the average preference probability of the ink volume adjustment images with the same maximum ink volume, so as to obtain the average preference probability of each ink volume limit image. The N1+N2+N3 ink volume adjustment images generated from the ink volume limitation image with the highest average preference probability are used as the target image set, and the skin color in the target image set is used as the preferred skin color for printed figures. Based on the ink volume of the target image, adjust the ink volume combination and preference probability of the image, and determine the amount of ink reduction (K). The images to be evaluated with the highest probability of preference, ranked T, are used as preferred color images, and a chromaticity scatter plot is drawn based on the chromaticity values ​​of the preferred color images; T represents the average number of images preferred by each observer when they judge whether they like the images to be evaluated. Based on the chromaticity scatter plot, fit the preferred ellipses of different color biases, and reduce the major and minor axes of the preferred ellipses to obtain the preferred reduced ellipses. Obtain the preference level of the samples on the preference ellipse and the reduced preference ellipse, and determine the chromaticity value of each sample on the preference ellipse and the reduced preference ellipse; the sample is a point on the chromaticity scatter plot corresponding to the center of the preference ellipse, the endpoints of the major and minor axes, and the endpoints of the major and minor axes of the reduced preference ellipse. Based on the preference level and the chromaticity values ​​of each sample, the C ink reduction, M ink reduction, and Y ink reduction of the target image set ink volume adjustment image are determined; the preferred skin tone of the printed person, the K ink reduction, the C ink reduction, the M ink reduction, and the Y ink reduction are used to print the preferred skin tone image.

2. The method for determining preferred skin tone and adjusting ink volume in printed portraits according to claim 1, characterized in that, For any ink volume limitation image, the ink volume combinations of the single-channel, dual-channel, and three-channel ink volume limitation images are adjusted respectively to generate N1+N2+N3 ink volume adjustment images, specifically including: For any ink-limited image, the ink volume of each single channel of the ink-limited image is reduced by a different proportion to obtain N1 single-channel ink-adjustment images; the single channel is one of the C channel, M channel, Y channel and K channel; For any ink-limited image, the ink volume is reduced by a different proportion in each dual channel of the ink-limited image to obtain N2 dual-channel ink-adjusted images; the dual channels are two of the C channel, M channel, Y channel and K channel; For any ink-limited image, reduce the ink volume of the K channel of the ink-limited image by a fixed proportion, and at the same time reduce the ink volume of both the C channel, M channel and Y channel of the ink-limited image by different proportions, to obtain N3 three-channel ink-limited adjustment images.

3. The method for determining preferred skin tone and adjusting ink volume in printed portraits according to claim 1, characterized in that, Clustering is performed on the skin color within the skin color region to obtain the clustering results, specifically including: An improved K-means algorithm is used to calculate the weighted Euclidean distance from each pixel in the skin color region to the cluster center; the improved K-means algorithm is a K-means algorithm with added weights; the weights are determined based on the contour coefficients of the pixels in the skin color region; The weighted Euclidean distance is used to determine the skin color category of each pixel in the skin color region, and the clustering result is obtained.

4. The method for determining preferred skin tone and adjusting ink volume in printed portraits according to claim 1, characterized in that, Based on the ink volume combination and preference probability of the target image set, determine the K ink reduction amount, specifically including: Compare the difference in preference probabilities between the ink amount adjustment images with reduced ink amount (K inks) and the ink amount adjustment images without reduced ink amount (K inks) in the target image set. If the difference is within the set difference range, then K ink needs to be reduced, and the value of the amount of K ink reduction needs to be determined. If the difference exceeds the set difference range, then there is no need to reduce K ink, and the amount of K ink reduction is zero.

5. The method for determining preferred skin tone and adjusting ink volume in printed portraits according to claim 1, characterized in that, Based on the chromaticity scatter plot, fit preference ellipses for different color biases, and reduce the major and minor axes of the preference ellipses to obtain preference-reduced ellipses, specifically including: Based on the chromaticity scatter plot, fit the whiter preference ellipse, the yellower preference ellipse, and the redder preference ellipse. Reduce the major and minor axes of the whitish preference ellipse to half their original size to obtain a whitish preference minimized ellipse. Reduce the major and minor axes of the yellowish preference ellipse to half their original size to obtain a yellowish preference minimized ellipse. Reduce the major and minor axes of the reddish preference ellipse to half their original size to obtain a reddish preference minimized ellipse.

6. The method for determining preferred skin tone and adjusting ink volume in printed portraits according to claim 3, characterized in that, The formula for calculating the weighted Euclidean distance is: ； in, Represents pixels in the skin color area a Weighted Euclidean distance to the cluster center; Represents pixels in the skin color area a The brightness of the color; Represents pixels in the skin color area a The saturation of the color; Represents pixels in the skin color area a The hue of the color; The brightness of the color representing the cluster centers; The saturation of the color representing the cluster centers; The hue of the color representing the cluster centers; m The weight for the difference in saturation; n Weights for hue differences; and It is based on pixels in the skin color region a It is calculated from the red-green and blue-yellow hues; and It is calculated based on the red-green and blue-yellow hues of the cluster centers.

7. The method for determining skin tone preferences and adjusting ink volume in printed portraits according to claim 1, characterized in that, Before calculating the average chromaticity values ​​of all feature color patches to obtain the chromaticity values ​​of the image to be evaluated, the process further includes: The chromaticity values ​​of each characteristic color patch were measured using a spectrophotometer with a geometric measurement condition of 45 / 0; the chromaticity values ​​included at least: red-green chromaticity and blue-yellow chromaticity.

8. The method for determining preferred skin tone and adjusting ink volume in printed portraits according to claim 1, characterized in that, S≧4; H≧2; The maximum ink consumption range is [260%, 340%].

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method for determining the preferred skin tone and adjusting the ink volume of a printed person as described in any one of claims 1-8.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method for determining the preferred skin tone and adjusting the ink volume of printed figures as described in any one of claims 1-8.

Images