A clothing color preference degree prediction method and system

Abstract

The application discloses a kind of clothing color preference prediction method and system, including the measurement of the spectral reflectance of the clothing to be evaluated;Calculate the color appearance attribute information of the clothing to be evaluated;The collection of the spectral reflectance of the face skin of the wearer of the clothing to be evaluated;Calculate the chroma attribute of the face skin of the wearer of the clothing to be evaluated;Calculate the perceived whiteness angle of the face skin of the wearer of the clothing to be evaluated;Determine whether the perceived whiteness angle of the face skin of the wearer of the clothing to be evaluated is within the scope applicable to the application;Collect the observation and evaluator gender of the clothing preference attribute to be evaluated;For the clothing to be evaluated, according to its color appearance information and evaluator gender, combined with clothing color preference quantification model, obtain the corresponding estimated value, realize the characterization of the preference degree of different color appearance attribute clothing.

Description

Technical Field

[0001] This invention belongs to the field of clothing color application, specifically relating to a method and system for predicting clothing color preferences. Background Technology

[0002] Because clothing color can affect a wearer's perception of skin tone, physical attractiveness, and body shape evaluation, people will show different degrees of preference for different colors of clothing. This attribute is called clothing color preference.

[0003] Existing research has shown that there are complex differences in color vision between men and women at the neuroscience and cognitive-behavioral levels. Men and women differ in color discrimination ability, color preference, color memory, and color recognition. Therefore, it is necessary to consider the gender and corresponding color vision characteristics of the observers and evaluators to accurately quantify and predict the degree of preference for different colors of clothing.

[0004] References: 1.Fink,B.,Grammer,K.,and Matts,P.(2006).Visible skin colordistribution plays a role in the perception of age,attractiveness,and healthin female faces.Evolution and Human Behavior 27,433-442.

[0005] 2.Kowal, M., Sorokowski, P., Pisanski, K., Valentova, JV, Varella, MAC, Frederick, DA, Al-Shawaf, L., García, FE, Giammusso, I., Gjoneska, B., et al. (2022). Predictors of enhancing human physical attractiveness: Data from 93 countries. Evolution and Human Behavior 43,455-474.

[0006] 3.Yund,EW,and Armington,JC(1975).Color and brightness contrast effects as a function of spatial variables.Vision Res 15,917-929.

[0007] 4.de Vries,GJ,and Forger,NG(2015).Sex differences in the brain:whole body perspective.Biology of Sex Differences 6,15.

[0008] However, current research by scholars both at home and abroad has not fully considered the influence of clothing color appearance attributes on clothing color preference. Furthermore, existing quantitative prediction methods for clothing preference do not consider the differences between men and women in color visual perception, nor do they fully consider the interactive effect between clothing color appearance attributes and the gender of the evaluator.

[0009] To address the aforementioned issues, there is an urgent need to propose a technical solution that addresses the impact of clothing color attributes on clothing preference, and to develop a quantitative prediction method and system for clothing color preference that aligns with human subjective color visual perception, while also taking into account the interactive effects between clothing color attributes and the gender of the evaluator. Summary of the Invention

[0010] The purpose of this invention is to solve the problems described in the background art by proposing a method and system for predicting clothing color preferences.

[0011] The technical solution of this invention is to provide a method for predicting clothing color preferences, comprising the following steps:

[0012] Step 1: Measure the spectral reflectance of the garment to be evaluated;

[0013] Step 2: Calculate the color appearance attribute information of the garment to be evaluated under a standard light source in the uniform color space S1;

[0014] Step 3: Measure the spectral reflectance of the facial skin of the wearer of the clothing to be evaluated;

[0015] Step 4: Calculate the facial skin color attribute L of the wearer of the clothing to be evaluated in color space S2. * a * and b * ;

[0016] Step 5, based on the facial skin color attribute L of the wearer of the clothing to be evaluated. * a * and b * Calculate the angle of perceived whiteness of facial skin tone.

[0017] Step 6: Determine the angle of perceived whiteness of the facial skin of the wearer of the clothing to be evaluated. Whether it is within the set range, i.e., judgment Check if the condition is met. If not, exit; if met, proceed to the next step.

[0018] Step 7: Collect the observation and evaluator's gender (sex) for the clothing preference attribute to be evaluated, with a value of 1 or 2.

[0019] Step 8: Combining the color and appearance attribute information of the clothing to be evaluated in Steps 2 and 7 with the evaluator's gender (sex) to construct the evaluation clothing color preference measurement model F, the preference estimate of the clothing to be evaluated is obtained, thereby realizing the representation of the degree of preference for clothing with different color and appearance attributes.

[0020] A quantitative model for clothing color preference, in the following form:

[0021] F = 2.00 * 10 -2 *J+2.10*10 -2 *a-5.10*10 -2 *b-1.42*10 -1 *M+6.00

[0022] *10 -3 *h+15.8*sex+2.00*10 -3 *sex*J-6.00*x10 -3 *sex

[0023] *a+3.00*10 -3 *sex*b-1.60*10 -2 *sex*M-5.65*10 -4

[0024] *sex*h-1.51*10 -4 *J 2 +4.50*10 -2 *a 2 +4.60*10 -2 *b 2

[0025] -4.20*10 -2 *M 2 -2.35*10 -5 *h 22 -5.11*sex 2 +5.62

[0026] Where F is the estimated value of color preference for the clothing, J is the brightness of the clothing to be evaluated, a is the red-green component of the clothing to be evaluated, b is the yellow-blue component of the clothing to be evaluated, M is the chroma of the clothing to be evaluated, h is the hue angle of the clothing to be evaluated, and sex is the gender of the evaluator.

[0027] Furthermore, the spectral reflectance of the clothing to be evaluated and the facial skin of the wearer of the clothing to be evaluated was measured using information in the 400nm-700nm band.

[0028] Furthermore, the uniform color space S1 adopts the CIE CAM16-UCS uniform color space; the standard light source is D65.

[0029] Furthermore, the uniform color space S2 adopts CIEL. * a * b * Color space.

[0030] Furthermore, the angle of perceived whiteness of facial skin tone. The specific calculation formula is as follows;

[0031]

[0032] ArcTan is the arctangent function.

[0033] This invention also provides a clothing color preference prediction system, comprising the following modules:

[0034] The apparel spectral reflectance measurement module is used to measure the spectral reflectance of the apparel to be evaluated.

[0035] The clothing color appearance attribute information calculation module is used to calculate the color appearance attribute information of the clothing to be evaluated under a standard light source in a uniform color space S1.

[0036] The skin spectral reflectance measurement module is used to measure the facial skin spectral reflectance of the wearer of the clothing being evaluated.

[0037] The skin tone attribute calculation module is used to calculate the facial skin tone attribute L of the wearer of the clothing being evaluated in the color space S2. * a * and b * ;

[0038] The facial skin tone perception whiteness angle calculation module is used to calculate the whiteness attribute L of the facial skin tone of the wearer of the clothing being evaluated. * a * and b * Calculate the angle of perceived whiteness of facial skin tone.

[0039] The judgment module is used to determine the perceived whiteness angle of the facial skin of the wearer of the clothing to be evaluated. Whether it is within the set range, i.e., judgment Check if it is valid; if not, exit; if valid, proceed to the next module.

[0040] The gender collection module is used to collect the gender of the observers and evaluators of the clothing preference attributes to be evaluated, with a value of 1 or 2.

[0041] The preference degree representation module is used to combine the color and appearance attribute information of the clothing to be evaluated and the color preference measurement model F of the clothing to be evaluated constructed by the evaluator's gender to obtain the preference degree estimate of the clothing to be evaluated, and thus realize the representation of the preference degree of clothing with different color and appearance attributes.

[0042] A quantitative model for clothing color preference, in the following form:

[0043] F = 2.00 * 10 -2 *J+2.10*10 -2 *a-5.10*10 -2 *b-1.42*10 -1 *M+6.00

[0044] *10 -3 *h+15.8*sex+2.00*10 -3 *sex*J-6.00*10 -3 *sex

[0045] *a+3.00*10 -3 *sex*b-1.60*10 -2 *sex*M-5.65*10 -4

[0046] *sex*h-1.51*10 -4 *J 2 +4.50*10 -2 *a 2 +4.60*10 -2 *b 2

[0047] -4.20*10 -2 *M 2 -2.35*10 -5 *h 2 -5.11*sex 2 +5.62

[0048] Where F is the estimated value of color preference for the clothing, J is the brightness of the clothing to be evaluated, a is the red-green component of the clothing to be evaluated, b is the yellow-blue component of the clothing to be evaluated, M is the chroma of the clothing to be evaluated, h is the hue angle of the clothing to be evaluated, and sex is the gender of the evaluator.

[0049] Furthermore, the spectral reflectance of the clothing to be evaluated and the facial skin of the wearer of the clothing to be evaluated was measured using information in the 400nm-700nm band.

[0050] Furthermore, the uniform color space S1 adopts the CIE CAM16-UCS uniform color space; the standard light source is D65.

[0051] Furthermore, the uniform color space S2 adopts CIEL. * a * b * Color space.

[0052] Furthermore, the angle of perceived whiteness of facial skin tone. The specific calculation formula is as follows;

[0053]

[0054] ArcTan is the arctangent function.

[0055] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0056] The present invention proposes a quantitative prediction technology for clothing color preferences. Based on the color characteristics of the clothing to be evaluated and using a quantitative model of clothing color preferences, it achieves a comprehensive and accurate representation of the degree of preference for clothing with different color attributes, thereby providing an effective and targeted method for evaluating the color perception characteristics of clothing in this field. Attached Figure Description

[0057] Figure 1 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0058] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0059] like Figure 1 The embodiment shown provides a clothing color preference quantification prediction technology solution. Based on the color characteristics of the clothing to be evaluated and using a clothing color preference quantification model, it achieves a comprehensive and accurate characterization of the preference degree for clothing with different color attributes, thereby providing an effective and targeted method for evaluating clothing color perception characteristics in this field.

[0060] The example uses 22 solid-color sweatshirts with different chromaticity attributes as the garments to be evaluated: the lightness (J) ranges from 15.74 to 90.36, the red-green component (a) ranges from -14.44 to 35.52, the yellow-blue component (b) ranges from -17.04 to 27.13, the saturation (M) ranges from 1.18 to 37.99, and the hue (h) range includes all hue angles. Four models with typical East Asian skin tones were used as the wearers of the garments to be evaluated. The accuracy of the proposed method for predicting clothing color preferences was demonstrated using psychophysical experimental results as the model validation basis. It should be noted that this invention is not limited to the aforementioned garments and skin tones; this method is also applicable to garments with other chromaticity attributes or models with different skin tones.

[0061] In specific implementations, the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. The method flow provided in the embodiments includes the following steps:

[0062] 1) Measure the spectral reflectance of the garment to be evaluated using information in the 400nm-700nm band;

[0063] In this embodiment, 22 solid-color sweatshirts with different chromaticity attributes were used as garments to be evaluated, and the spectral reflectance of the garments to be evaluated was measured using an X-Rite i1 Pro spectrophotometer with a wavelength range of 400nm-700nm.

[0064] 2) Calculate the color appearance attribute information of the garment to be evaluated under the standard light source D65 in the uniform color space S1;

[0065] In this embodiment, the CAM16-UCS color space is used to calculate the hue attribute information of the garment to be evaluated under the D65 standard light source. For the specific method of calculating the hue attribute information in the CAM16-UCS color space, please refer to Li C, Li Z, Wang Z, et al. Comprehensive color solutions: CAM16, CAT16, and CAM16-UCS[J]. Color Research & Application, 2017, 42(6):703-718. This invention will not elaborate further.

[0066] 3) Measure the spectral reflectance of the facial skin of the wearer of the clothing to be evaluated, using information in the 400nm-700nm band;

[0067] In this embodiment, the spectral reflectance of the facial skin of the wearer of the clothing to be evaluated was measured using an X-Rite i1 Pro spectrophotometer in the wavelength range of 400nm-700nm.

[0068] 4) Calculate the facial skin color attribute L of the wearer of the clothing to be evaluated in the color space S2. * a * and b * ;

[0069] In the embodiment, CIEL is used. * a * b * Color space, calculate the facial skin color attributes of the wearer of the clothing to be evaluated under the D65 standard light source.

[0070] 5) The facial skin color attribute L of the wearer of the clothing to be evaluated * a * and b * Inputting the angle of perceived whiteness of facial skin tone The specific calculation formula is as follows;

[0071]

[0072] In the embodiment, the perceived whiteness angles of facial skin tone for the four clothing wearers to be evaluated were as follows:

[0073] 6) Determine the angle of perceived whiteness of the facial skin of the wearer of the clothing to be evaluated. Whether it falls within the scope of this invention, i.e., determining Whether it is valid or not, if it is not valid, then this invention is not applicable; if it is valid, then proceed to the next step.

[0074] In the embodiment, a = 0, b = 60.

[0075] 7) Collect observations and evaluators' gender (sex) on the clothing preference attributes to be evaluated;

[0076] In this embodiment, if the evaluator is male, the gender is assigned the value sex=1; if the evaluator is female, the gender is assigned the value sex=2.

[0077] 8) Input the color and appearance attribute information of the clothing to be evaluated and the gender (sex) of the evaluator into the color preference quantification model F constructed in this invention to obtain the preference estimate of the clothing to be evaluated, thereby realizing the representation of the preference degree for clothing with different color and appearance attributes.

[0078] F is a quantitative model for the degree of preference for clothing colors, and its specific form is as follows:

[0079] F = 2.00 * 10 -2 *J+2.10*10 -2 *a-5.10*10 -2 *b-1.42*10 -1*M+6.00

[0080] *10 -3 *h+15.8*sex+2.00*10 -3 *sex*J-6.00*10 -3 *sex

[0081] *a+3.00*10 -3 *sex*b-1.60*10 -2 *sex*M-5.65*10 -4

[0082] *sex*h-1.51*10 -4 *J 2 +4.50*10 -2 *a 2 +4.60*10 -2 *b 2

[0083] -4.20*10 -2 *M 2 -2.35*10 -5 *h 2 -5.11*sex 2 +5.62

[0084] Where F is the estimated value of color preference for the clothing, J is the brightness of the clothing to be evaluated, a is the red-green component of the clothing to be evaluated, b is the yellow-blue component of the clothing to be evaluated, M is the chroma of the clothing to be evaluated, h is the hue angle of the clothing to be evaluated, and sex is the gender of the evaluator.

[0085] To further demonstrate the technical advantages of the method described in this invention in evaluating clothing color preferences, a subjective comparative experiment was conducted, and the goodness of fit R was measured. 2 Using the correlation coefficient R method, the goodness of fit R between the subjective evaluation values ​​of observers' preferences for clothing obtained from the subjective experiment and the clothing color preference quantification model F in (8) is calculated. 2 And the correlation coefficient R. The specific implementation was as follows: using the aforementioned 22 solid-color sweatshirts with different chromaticity attributes as the garments to be evaluated, and four models with typical East Asian skin tones as the wearers of the garments to be evaluated, a subjective evaluation experiment of preference was conducted. The specific experimental method is as follows:

[0086] 1) The experimental lighting was constructed using two LightCubes. The two cubes were positioned at a 120° angle to each other on the left and right sides of the experimental scene, tilted at a 30° angle to illuminate the black chair in the experimental area. The height of the black chair was adjustable; before the experiment, the chair height was adjusted to ensure even illumination of the model's face (measurements showed that the illuminance on both sides of the model's face was similar). The correlated color temperature (CCT) of the light source was 6500K, and the illuminance was 750 lx. During the experiment, the average illuminance on the side and front of the model's face was 82 lx.

[0087] 2) A total of 66 observers (33 males and 33 females) participated in the subjective evaluation experiment on preferences. The youngest participant was 18 years old, the oldest was 38 years old, and the average age was 22 years old. Before the experiment officially began, the observers were required to take the Ishihara color blindness test to ensure that they did not have any visual impairment in color perception. After passing the test, they filled out a questionnaire outside the laboratory containing information such as gender, age, and skin color. When the observers entered the laboratory, they changed into a gray lab coat to prevent problems such as glare from the clothing and the clothing's own color interfering with their judgment, and were asked to turn off their mobile phones and other light-emitting devices.

[0088] 3) In the subjective evaluation experiment of clothing color preference, the observers made subjective judgments and ratings on the degree of preference for 22 pieces of clothing in a random order.

[0089] The above subjective experiments yielded the observers' subjective evaluations of their preferences for the 22 garments, and further calculated the goodness of fit R between these evaluations and the garment color preference quantification model F constructed in this invention. 2 The correlation coefficients R are shown in Table 1. The results show that the goodness of fit between the subjective evaluation value and the model estimate is 0.798 and the correlation coefficient is 0.893, respectively, proving that the clothing color preference quantification model F constructed in this invention has extremely high accuracy, and further demonstrating that the method described in this invention has strong technical advantages in evaluating clothing color preferences.

[0090] Table 1. Goodness of fit R between subjective evaluation values ​​and model estimates 2 and correlation coefficient R

[0091] Preference 0.798 0.893

[0092] On the other hand, embodiments of the present invention also provide a clothing color preference prediction system, including the following modules:

[0093] The apparel spectral reflectance measurement module is used to measure the spectral reflectance of the apparel to be evaluated.

[0094] The clothing color appearance attribute information calculation module is used to calculate the color appearance attribute information of the clothing to be evaluated under a standard light source in a uniform color space S1.

[0095] The skin spectral reflectance measurement module is used to measure the facial skin spectral reflectance of the wearer of the clothing being evaluated.

[0096] The skin tone attribute calculation module is used to calculate the facial skin tone attribute L of the wearer of the clothing being evaluated in the color space S2. * a * and b * ;

[0097] The facial skin tone perception whiteness angle calculation module is used to calculate the whiteness attribute L of the facial skin tone of the wearer of the clothing being evaluated. * a * and b * Calculate the angle of perceived whiteness of facial skin tone.

[0098] The judgment module is used to determine the perceived whiteness angle of the facial skin of the wearer of the clothing to be evaluated. Whether it is within the set range, i.e., judgment Check if it is valid; if not, exit; if valid, proceed to the next module.

[0099] The gender collection module is used to collect the gender of the observers and evaluators of the clothing preference attributes to be evaluated, with a value of 1 or 2.

[0100] The preference degree representation module is used to combine the color and appearance attribute information of the clothing to be evaluated and the color preference measurement model F of the clothing to be evaluated constructed by the evaluator's gender to obtain the preference degree estimate of the clothing to be evaluated, and thus realize the representation of the preference degree of clothing with different color and appearance attributes.

[0101] A quantitative model for clothing color preference, in the following form:

[0102] F = 2.00 * 10 -2 *J+2.10*10 -2 *a-5.10*10 -2 *b-1.42*10 -1 *M+6.00

[0103] *10 -3 *h+15.8*sex+2.00*10 -3 *sex*J-6.00*10 -3 *sex

[0104] *a+3.00*10 -3 *sex*b-1.60*10 -2 *sex*M-5.65*10 -4

[0105] *sex*h-1.51*10 -4 *J 2 +4.50*10 -2 *a 2 +4.60*10 -2 *b 2

[0106] -4.20*10 -2 *M 2 -2.35*10 -5 *h 2 -5.11*sex 2 +5.62

[0107] Where F is the estimated value of color preference for the clothing, J is the brightness of the clothing to be evaluated, a is the red-green component of the clothing to be evaluated, b is the yellow-blue component of the clothing to be evaluated, M is the chroma of the clothing to be evaluated, h is the hue angle of the clothing to be evaluated, and sex is the gender of the evaluator.

[0108] The specific implementation methods of each module are the same as those of each step, and will not be described in this invention.

[0109] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A method for predicting clothing color preferences, characterized in that, Includes the following steps: Step 1: Measure the spectral reflectance of the garment to be evaluated; Step 2: Calculate the color appearance attribute information of the garment to be evaluated under standard light source in the uniform color space S1; Step 3: Measure the spectral reflectance of the facial skin of the wearer of the clothing to be evaluated; Step 4: Calculate the facial skin color attribute L of the wearer of the clothing to be evaluated in color space S2. * a * and b * ; Step 5, based on the facial skin color attribute L of the wearer of the clothing to be evaluated. * a * and b * Calculate the perceived whiteness angle of facial skin tone. Step 6: Determine the angle of perceived whiteness of the facial skin of the wearer of the clothing to be evaluated. Whether it is within the set range, i.e., judgment Check if the condition is met. If not, exit; if met, proceed to the next step. Step 7: Collect the observation and evaluator's gender (sex) for the clothing preference attribute to be evaluated, with a value of 1 or 2. Step 8: Combining the color and appearance attribute information of the clothing to be evaluated in Steps 2 and 7 with the evaluator's gender (sex) to construct the evaluation clothing color preference measurement model F, the preference estimate of the clothing to be evaluated is obtained, thereby realizing the representation of the degree of preference for clothing with different color and appearance attributes. A quantitative model for clothing color preference, in the following form: F=2.00*10 -2 *J+2.10*10 -2 *a-5.10*10 -2 *b-1.42*10 -1 *M+6.00*10 -3 *h+15.8*sex+2.00*10 -3 *sex*J-6.00*10 -3 *sex*a+3.00*10 -3 *sex*b-1.60*10 -2 *sex*M-5.65*10 -4 *sex*h-1.51*10 -4 *J 2 +4.50*10 -2 *a 2 +4.60*10 -2 *b 2 -4.20*10 -2 *M 2 -2.35*10 -5 *h 2 -5.11*sex 2 +5.62 Where F is the estimated value of color preference for the clothing, J is the brightness of the clothing to be evaluated, a is the red-green component of the clothing to be evaluated, b is the yellow-blue component of the clothing to be evaluated, M is the chroma of the clothing to be evaluated, h is the hue angle of the clothing to be evaluated, and sex is the gender of the evaluator.

2. The method for predicting clothing color preferences as described in claim 1, characterized in that: The spectral reflectance of the clothing to be evaluated and the facial skin of the wearer of the clothing to be evaluated was measured using information in the 400nm-700nm band.

3. The method for predicting clothing color preferences as described in claim 1, characterized in that: The uniform color space S1 adopts the CIE CAM16-UCS uniform color space; the standard light source is D65.

4. The method for predicting clothing color preferences as described in claim 1, characterized in that: Uniform color space S2 adopts CIE L * a * b * Color space.

5. The method for predicting clothing color preferences as described in claim 1, characterized in that: Facial skin tone perception whiteness angle The specific calculation formula is as follows; ArcTan is the arctangent function.

6. A clothing color preference prediction system, characterized in that, Includes the following modules: The apparel spectral reflectance measurement module is used to measure the spectral reflectance of the apparel to be evaluated. The clothing color appearance attribute information calculation module is used to calculate the color appearance attribute information of the clothing to be evaluated under a standard light source in a uniform color space S1. The skin spectral reflectance measurement module is used to measure the facial skin spectral reflectance of the wearer of the clothing being evaluated. The skin tone attribute calculation module is used to calculate the facial skin tone attribute L of the wearer of the clothing being evaluated in the color space S2. * a * and b * ; The facial skin tone perception whiteness angle calculation module is used to calculate the whiteness attribute L of the facial skin tone of the wearer of the clothing being evaluated. * a * and b * Calculate the perceived whiteness angle of facial skin tone. The judgment module is used to determine the perceived whiteness angle of the facial skin of the wearer of the clothing to be evaluated. Whether it is within the set range, i.e., judgment Check if it is valid; if not, exit; if valid, proceed to the next module. The gender collection module is used to collect the gender of the observers and evaluators of the clothing preference attributes to be evaluated, with a value of 1 or 2. The preference degree representation module is used to combine the color and appearance attribute information of the clothing to be evaluated and the color preference measurement model F of the clothing to be evaluated constructed by the evaluator's gender to obtain the preference degree estimate of the clothing to be evaluated, and thus realize the representation of the preference degree of clothing with different color and appearance attributes. A quantitative model for clothing color preference, in the following form: F=2.00*10 -2 *J+2.10*10 -2 *a-5.10*10 -2 *b-1.42*10 -1 *M+6.00*10 -3 *h+15.8*sex+2.00*10 -3 *sex*J-6.00*10 -3 *sex*a+3.00*10 -3 *sex*b-1.60*10 -2 *sex*M-5.65*10 -4 *sex*h-1.51*10 -4 *J 2 +4.50*10 -2 *a 2 +4.60*10 -2 *b 2 -4.20*10 -2 *M 2 -2.35*10 -5 *h 2 -5.11*sex 2 +5.62 Where F is the estimated value of color preference for the clothing, J is the brightness of the clothing to be evaluated, a is the red-green component of the clothing to be evaluated, b is the yellow-blue component of the clothing to be evaluated, M is the chroma of the clothing to be evaluated, h is the hue angle of the clothing to be evaluated, and sex is the gender of the evaluator.

7. The clothing color preference prediction system as described in claim 6, characterized in that: The spectral reflectance of the clothing to be evaluated and the facial skin of the wearer of the clothing to be evaluated was measured using information in the 400nm-700nm band.

8. The clothing color preference prediction system as described in claim 6, characterized in that: The uniform color space S1 adopts the CIE CAM16-UCS uniform color space; the standard light source is D65.

9. The method for predicting clothing color preferences as described in claim 6, characterized in that: Uniform color space S2 adopts CIE L * a * b * Color space.

10. The method for predicting clothing color preferences as described in claim 6, characterized in that: Facial skin tone perception whiteness angle The specific calculation formula is as follows; ArcTan is the arctangent function.

Images