Textile color migration detection system and method

Abstract

The invention relates to the technical field of textile detection, and discloses a textile color migration detection system and method. According to the system and the method, textile image data are acquired through a high-precision scanner, an image processing module sequentially performs noise removal, brightness adjustment and contrast enhancement operation on the data, and a color space conversion module converts an image from an RGB color space to a Lab color space based on a contrast-enhanced textile image. The color feature extraction module extracts color features of the textile image, the color features comprise Euclidean distances between a color histogram and colors, calculates a standard value of differences between the colors and transmits the standard value to the migration detection module, and the migration detection module sets a color migration threshold value; and the color migration detection module is used for comparing with a standard value of color difference to judge whether the color of the textile is migrated or not, generating a color migration detection report and transmitting the color migration detection report to the visualization module, and the visualization module displays the color migration detection report on a visualization terminal.

Description

Technical Field

[0001] The invention relates to the technical field of textile detection, and in particular to a textile color migration detection system and method. Background Art

[0002] The textile color migration detection system is a technology used to monitor and evaluate the color change or migration of textiles during production, processing, storage and use. This technology is mainly used in the textile and garment industry, especially in dyeing, printing, washing and other post-processing processes to ensure the color quality and consistency of textiles. The system can help manufacturers ensure the color consistency of each batch of dyed or printed products, reduce the risk of product returns due to color inconsistency, and detect color migration or changes in a timely manner, which helps to detect problems as early as possible and take measures, and can reduce the waste of raw materials and time caused by re-dyeing or other corrections.

[0003] Currently, most color testing of textiles is done by professionals who rely on their own profound experience to conduct color analysis. However, for complex color mixing, gradient or printing effects, it is difficult to provide accurate assessments based solely on the professionals' own experience. Therefore, it is necessary to introduce high-end color testing equipment for testing, which will increase operating costs for textile companies. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a textile color migration detection system and method, which are capable of acquiring textile image data Fztx through a high-precision scanner, and transmitting the acquired data to an image processing module through a network, wherein the image processing module sequentially performs noise removal, brightness adjustment, and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz, and a contrast enhanced textile image Dbzq, a color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq for analyzing the color change of the textile, and a color feature extraction module extracts the textile image according to the converted Lab color space. The color features of the textile are obtained by the color detection module, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and the standard numerical value ΔE of the difference between the colors is calculated and the value is transmitted to the migration detection module. The migration detection module sets the color migration threshold Qyyz for comparison with the standard numerical value ΔE of the difference between the colors to determine whether the color of the textile has migrated, and generates a color migration detection report and transmits it to the visualization module. The visualization module displays the color migration detection report on the visualization terminal, calculates the relevant parameters of the textile image, sets the threshold for color migration detection, does not require manual judgment, and does not need to introduce high-end color detection equipment to accurately detect the color migration of complex color mixing, gradient or printing of textiles, reduces the operating costs of textile enterprises, and solves the above problems.

[0006] (II) Technical solution

[0007] To achieve the above-mentioned object, the present invention provides the following technical solutions: a textile color migration detection system, comprising an image acquisition module, an image processing module, a color space conversion module, a color feature extraction module, a migration detection module and a visualization module;

[0008] The image acquisition module acquires textile image data Yztx through a high-precision scanner, and transmits the acquired data to the image processing module through a network;

[0009] The image processing module sequentially performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq, and transmits the contrast enhanced textile image Dbzq to the color space conversion module through a network;

[0010] The color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, so as to analyze the color change of the textile;

[0011] The color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and calculates the standard value ΔE of the difference between the colors and transmits the value to the migration detection module;

[0012] The migration detection module sets a color migration threshold Qyyz for comparison with a standard value ΔE of the difference between colors to determine whether the color of the textile has migrated, and generates a color migration detection report to be transmitted to the visualization module;

[0013] The visualization module displays the color migration detection report on the visualization terminal.

[0014] Preferably, the image processing module removes noise from the textile image data Fztx transmitted by the image acquisition module to obtain a denoised textile image Qzfz. The denoising formula is as follows:

[0015]

[0016] In the formula, Qzfz(x,y) represents the pixel value of the denoised textile image, Fztx(j,j) represents the pixel value in the area centered on (x,y) of the original textile image data, K represents the total number of pixels in the area, S represents the set of pixels in the area, and (i,j) represents the pixel point of the original textile image data.

[0017] Preferably, the image processing module performs a brightness adjustment operation on the textile image data Fztx transmitted by the image acquisition module to obtain a brightness adjusted textile image Ldtz. The brightness adjustment formula is as follows:

[0018] Ldtz(x,y)=Qzfz(x,y)+C

[0019] In the formula, Ldtz(x,y) represents the pixel value of the textile image after brightness adjustment, Qzfz(x,y) represents the pixel value of the textile image after denoising, and C represents the adjusted brightness parameter, which is set by the system.

[0020] Preferably, the image processing module performs a contrast enhancement operation on the textile image data Fztx transmitted by the image acquisition module to obtain a contrast enhanced textile image Dbzq. The contrast enhancement formula is as follows:

[0021] Dbzq(x,y)=α*(Ldtz(x,y)-128)+128

[0022] In the formula, Dbzq(c,y) represents the contrast-enhanced textile image pixel value, Ldtz(x,y) represents the brightness-adjusted textile image pixel value, α represents the contrast enhancement factor, and 128 is a constant used to center the pixel value from 0 to 255 to ensure that the calculation can be performed within the appropriate range.

[0023] Preferably, the color space conversion module converts the image from the RGB color space to the Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, and the conversion formula is as follows:

[0024] To convert an image from RGB color space to Lab color space, you need to convert the RGB color space to XYZ color space first. The RGB values ​​of the image are R, G, B. First, you need to normalize the RGB values ​​to the [0,1] range, and then use the following formula to convert:

[0025]

[0026] Then use the linear conversion formula for conversion:

[0027] X=R'*0.4124564+G'*0.3575761+B'*0.1804375

[0028] Y=R'*0.2126729+G'*0.7151522+B'*0.0721750

[0029] Y=R'*0.0193339+G'*0.1191920+B'*0.9503041

[0030] Convert XYZ to Lab and calculate the normalized value X n , Y n , Z n , take the white point value, the values ​​are as follows:

[0031] X n =95.047

[0032] Y n =100.000

[0033] Z n =108.883

[0034] Use the following formula to convert:

[0035]

[0036] The function f(t) is defined as follows:

[0037]

[0038] In the above formula, R', G', B' represent the normalization of the R, G, B values ​​of the image, L, a, b represent the corresponding values ​​after the linear conversion of X, Y, Z, f(t) represents the conversion function, and t represents the parameter in the conversion function.

[0039] Preferably, the color feature extraction module extracts the color histogram Yzft of the textile image according to the converted Lod color space, and the calculation formula is as follows:

[0040]

[0041] In the formula, Yzft L Represents the color histogram index of the textile image of the L channel. L represents the L channel value of the Lab color space. The range is [0, 100]. Max L Indicates the maximum L value of the color channel, Yzft a The color histogram index of the textile image representing the a channel, Yzft b Represents the color histogram index of the textile image of the b channel, a represents the a channel value of the Lab color space, the range is [-128, 127], b represents the b channel value of the Lab color space, the range is [-128, 127], n L 、n a 、n b They respectively represent the number of intervals of channel L, the number of intervals of channel a, and the number of intervals of channel b, which are set by the system.

[0042] Preferably, the color feature extraction module extracts the color features of the textile image according to the converted Lab color space, and calculates the Euclidean distance Ojjl of the color, and the calculation formula is as follows:

[0043]

[0044] In the formula, Ojjl represents the Euclidean distance of colors, L1, a1, b1 represent the Lab value of the first color, and L2, a2, b2 represent the Lab value of the second color.

[0045] Preferably, the color feature extraction module extracts the color features of the textile image according to the converted Lab color space to calculate the standard value ΔE of the difference between colors, and the calculation formula is as follows:

[0046]

[0047] In the formula, ΔE represents the standard value of the difference between colors, and ΔL, Δa, and Δb represent the differences in brightness, red-green components, and blue-yellow components between two pixels in the Lab color space.

[0048] Preferably, the migration detection module sets a color migration threshold Qyyz for comparison with a standard value ΔE of the color difference to determine whether the color of the textile has migrated, and the comparison method is as follows:

[0049] When the standard value ΔE of the difference between colors exceeds the color migration threshold Qyyz set by the migration detection module, it indicates that the color of the textile has migrated.

[0050] A method for detecting color migration of textiles comprises the following steps:

[0051] S1, the image acquisition module obtains the textile image data Fztx through a high-precision scanner, and transmits the acquired data to the image processing module through the network;

[0052] S2, the image processing module performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module in sequence, to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq;

[0053] S3, the color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, which is used to analyze the color change of textiles;

[0054] S4, the color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and calculates the standard value ΔE of the difference between the colors and transmits the value to the migration detection module;

[0055] S5, the migration detection module sets a color migration threshold Qyyz for comparison with the standard value ΔE of the color difference to determine whether the color of the textile has migrated, and generates a color migration detection report to be transmitted to the visualization module;

[0056] S6. The visualization module displays the color migration detection report on the visualization terminal.

[0057] Compared with the prior art, the present invention provides a textile color migration detection system and method, which has the following beneficial effects:

[0058] The present invention acquires textile image data Fztx through a high-precision scanner, and transmits the acquired data to an image processing module through a network. The image processing module sequentially performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq. The color space conversion module converts the image from the RGB color space to the Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq for analyzing the color change of the textile. The color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including color The Euclidean distance Ojjl between the histogram Yzft and the color is calculated, and the standard numerical value ΔE of the difference between colors is calculated and transmitted to the migration detection module. The migration detection module sets the color migration threshold Oyyz, which is used to compare with the standard numerical value ΔE of the difference between colors to determine whether the color of the textile has migrated, and generates a color migration detection report and transmits it to the visualization module. The visualization module displays the color migration detection report on the visualization terminal, calculates the relevant parameters of the textile image, sets the threshold for color migration detection, and does not require manual judgment or the introduction of high-end color detection equipment to accurately detect the color migration of complex color mixing, gradient or printing of textiles, thereby reducing the operating costs of textile enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic diagram of the system flow of the present invention;

[0060] Figure 2 It is a schematic diagram of the steps of the method of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] At present, it is difficult to provide accurate evaluation of the complex color mixing, gradient or printing effects of textiles based on the experience of professionals. Therefore, it is necessary to introduce high-end color detection equipment for detection, which will increase the operating costs of textile companies. For this reason, a textile color migration detection system is proposed. Please refer to Figure 1 ,The system includes an image acquisition module, an image processing module, a color space conversion module, a color feature extraction module, a migration detection module and a visualization module;

[0063] The image acquisition module acquires textile image data Fztx through a high-precision scanner and compiles the acquired data into an image data set, which is {Fztx n-9 ,Fztx n-8 ,...,ztx n}, the above collected data is transmitted to the image processing module through the network;

[0064] The image processing module performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module in sequence to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq, where:

[0065] The image denoising formula is as follows:

[0066]

[0067] Textile image denoising is an important image processing step that aims to improve the quality and usability of images. In the process of color migration detection, textile defect detection and other analysis, clean images help improve the accuracy and reliability of measurement. In the formula, Qzfz(x,y) represents the pixel value of the denoised textile image, Fztx(i,j) represents the pixel value in the field centered on (x,y) of the original textile image data, K represents the total number of pixels in the field, S represents the set of pixels in the field, and (i,j) represents the pixel point of the original textile image data. Denoising can reduce random noise in the image, making the details of the textile more obvious and improving the clarity of the overall image.

[0068] The brightness adjustment formula is as follows:

[0069] Ldtz(x,y)=Qzfz(x,y)+C

[0070] Brightness adjustment can highlight the true color of the fabric, make it consistent under different lighting conditions, ensure color accuracy, and help eliminate color distortion caused by insufficient or excessive light. In the formula, Ldtz(x,y) represents the pixel value of the textile image after brightness adjustment, Qzfz(x,y) represents the pixel value of the textile image after denoising, and C represents the adjusted brightness parameter, which is set by the system. Consistent brightness makes the subsequent image feature extraction processing steps more efficient and reduces the unnecessary complexity caused by uneven lighting.

[0071] The contrast enhancement formula is as follows:

[0072] Dbzq(x,y)=α*(Ldtz(x,y)-128)+128

[0073] In the inspection process of textiles, enhanced contrast can help to more easily identify color unevenness, defects and other potential problems, thereby improving the accuracy of quality control. In the formula, Dbzq(x,y) represents the pixel value of the textile image with contrast enhancement, Ldtz(x,y) represents the pixel value of the textile image after brightness adjustment, α represents the contrast enhancement factor, and 128 is a constant used to center the pixel value from 0-255 to ensure that the calculation can be performed within the appropriate range. By enhancing the contrast, the key features and information of the image will be more prominent and easier to convey to the target audience, thereby improving the effectiveness of the information;

[0074] The color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, which is used to analyze the color change of textiles, where:

[0075] The color space conversion formula is as follows:

[0076] To convert an image from RGB color space to Lab color space, you need to convert the RGB color space to XYZ color space first. The RGB values ​​of the image are R, G, B. First, you need to normalize the RGB values ​​to the [0,1] range, and then use the following formula to convert:

[0077]

[0078] Then use the linear conversion formula for conversion:

[0079] X=R'*0.4124564+G'*0.3575761+B'*0.1804375

[0080] Y=R'*0.2126729+G'*0.7151522+B'*0.0721750

[0081] Y=R'*0.0193339+Y'*0.1191920+B'*0.9503041

[0082] Convert XYZ to Lab and calculate the normalized value X n , Y n , Z n , take the white point value, the values ​​are as follows:

[0083] X n =95.047

[0084] Y n =100.000

[0085] Z n =108.883

[0086] Use the following formula to convert:

[0087]

[0088] The function f(t) is defined as follows:

[0089]

[0090] In the above formula, R', G', B' represent the normalization of the R, G, B values ​​of the image, L, a, b represent the corresponding values ​​after the linear conversion of X, Y, Z, f(t) represents the conversion function, and t represents the parameter in the conversion function. Through the above conversion steps, the image can be converted from the RGB color space to the Lab color space;

[0091] The color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and calculates the standard value ΔE of the difference between the colors and transmits the value to the migration detection module;

[0092] The color histogram calculation formula is as follows:

[0093]

[0094] The color histogram can clearly show the frequency of each color in the image, making the analysis of color distribution in the image intuitive, especially useful in image feature extraction and visual analysis. In the formula, Yzft L Represents the color histogram index of the textile image of the L channel. L represents the L channel value of the Lab color space. The range is [0, 100]. Max L Indicates the maximum L value of the color channel, Yzft a The color histogram index of the textile image representing the a channel, Yzft b Represents the color histogram index of the textile image of the b channel, a represents the a channel value of the Lab color space, the range is [-128, 127], b represents the b channel value of the Lab color space, the range is [-128, 127], n L 、n a 、n b Respectively represent the interval number of channel L, the interval number of channel a, and the interval number of channel b, which are set by the system. By analyzing the color histogram, the color quality of the image can be evaluated, and problems such as color cast, overexposure or underexposure can be found, thereby providing a basis for post-processing and adjustment of the image;

[0095] The Euclidean distance calculation formula for color is as follows:

[0096]

[0097] The Euclidean distance can intuitively reflect the similarity between two colors. By calculating the distance between colors, we can quickly determine whether the colors are similar, so as to classify or match them. In the formula, Ojjl represents the Euclidean distance of colors, L1, a1, b1 represent the Lab value of the first color, and L2, a2, b2 represent the Lab value of the second color. In the process of converting between different color spaces, using the Euclidean distance can help evaluate the consistency of colors and ensure the similarity between the converted colors and the original colors.

[0098] The standard numerical value ΔE of the difference between colors is calculated as follows:

[0099]

[0100] ΔE is a standardized value used to quantify the difference between colors. By quantifying the difference between colors, the comparison between different colors becomes more scientific and systematic, and the color quality and consistency can be evaluated. In the formula, ΔE represents the standard value of the difference between colors, and ΔL, Δa, and Δb represent the difference in brightness, red-green components, and blue-yellow components between two pixels in the Lab color space. In the printing, coating, textile and other industries, ΔE can be used as a quality control tool to detect the deviation between the final product and the standard color to ensure that the color standards required by customers and the industry are met. At the same time, ΔE analysis can be used to identify color problems in the production process, thereby improving the process, reducing color difference, and reducing the scrap rate caused by color inconsistency.

[0101] The migration detection module sets a color migration threshold Qyyz, which is used to compare with the standard value ΔE of the difference between colors to determine whether the color of the textile has migrated. The comparison method is as follows:

[0102] When the standard value ΔE of the color difference exceeds the color migration threshold Qyyz set by the migration detection module, it indicates that the color of the textile has migrated. At the same time, the module generates a color migration detection report and transmits it to the visualization module. The visualization module displays the color migration detection report on the visualization terminal.

[0103] Please refer to Figure 2 , a method for detecting color migration of textiles, the method comprising the following steps:

[0104] S1, the image acquisition module obtains the textile image data Fztx through a high-precision scanner, and transmits the acquired data to the image processing module through the network;

[0105] S2, the image processing module performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module in sequence, to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq;

[0106] S3, the color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, which is used to analyze the color change of textiles;

[0107] S4, the color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and calculates the standard value ΔE of the difference between the colors and transmits the value to the migration detection module;

[0108] S5, the migration detection module sets a color migration threshold Qyyz for comparison with the standard value ΔE of the color difference to determine whether the color of the textile has migrated, and generates a color migration detection report to be transmitted to the visualization module;

[0109] S6. The visualization module displays the color migration detection report on the visualization terminal.

[0110] By combining the above-mentioned system and method, by calculating the relevant parameters of the textile image and setting the threshold for color migration detection, there is no need for manual judgment or the introduction of high-end color detection equipment to accurately detect the color migration of complex color mixing, gradient or printing of textiles, thereby reducing the operating costs of textile companies.

[0111] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A textile color migration detection system, characterized in that: It includes an image acquisition module, an image processing module, a color space conversion module, a color feature extraction module, a migration detection module and a visualization module; The image acquisition module acquires textile image data Fztx through a high-precision scanner, and transmits the acquired data to the image processing module through a network; The image processing module performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module in sequence, obtains a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq, and transmits the contrast enhanced textile image Dbzq to the color space conversion module through a network; The color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, so as to analyze the color change of the textile; The color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and calculates the standard value ΔE of the difference between the colors and transmits the value to the migration detection module; The migration detection module sets a color migration threshold Qyyz for comparison with a standard value ΔE of the difference between colors to determine whether the color of the textile has migrated, and generates a color migration detection report to be transmitted to the visualization module; The visualization module displays the color migration detection report on the visualization terminal.

2. A textile color migration detection system according to claim 1, characterized in that: The image processing module removes noise from the textile image data Fztx transmitted by the image acquisition module to obtain a denoised textile image Qzfz. The denoising formula is as follows: In the formula, Qzfz(x, y) represents the pixel value of the denoised textile image, Fztx(i, j) represents the pixel value in the area centered on (x, y) of the original textile image data, K represents the total number of pixels in the area, S represents the set of pixels in the area, and (i, j) represents the pixel point of the original textile image data.

3. A textile color migration detection system according to claim 2, characterized in that: The image processing module performs a brightness adjustment operation on the textile image data Fztx transmitted by the image acquisition module to obtain a brightness adjusted textile image Ldtz. The brightness adjustment formula is as follows: Ldtz(x,y)=Qzfz(x,y)+C In the formula, Ldtz(x, y) represents the pixel value of the textile image after brightness adjustment, Qzfz(x, y) represents the pixel value of the textile image after denoising, and C represents the adjusted brightness parameter, which is set by the system.

4. A textile color migration detection system according to claim 3, characterized in that: The image processing module performs a contrast enhancement operation on the textile image data Fztx transmitted by the image acquisition module to obtain a contrast enhanced textile image Dbzq. The contrast enhancement formula is as follows: Dbzq(x,y)=α*(Ldtz(x,y)-128)+128 In the formula, Dbzq(x, y) represents the contrast-enhanced textile image pixel value, Ldtz(x, y) represents the brightness-adjusted textile image pixel value, α represents the contrast enhancement factor, and 128 is a constant used to center the pixel value from 0 to 255 to ensure that the calculation can be performed within the appropriate range.

5. A textile color migration detection system according to claim 4, characterized in that: The color space conversion module converts the image from the RGB color space to the Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq. The conversion formula is as follows: To convert an image from RGB color space to Lab color space, you need to convert the RGB color space to XYZ color space first. The RGB values ​​of the image are R, G, B. First, you need to normalize the RGB values ​​to the range of [0, 1], and then use the following formula to convert: Then use the linear conversion formula for conversion: X=R′*0.4124564+G′*0.3575761+B′*0.1804375 Y=R′*0.2126729+G′*0.7151522+B′*0.0721750 Z=R′*0.0193339+G'*0.1191920+B'*0.9503041 Convert XYZ to Lab and calculate the normalized value X n , Y n , Z n , take the white point value, the values ​​are as follows: X n ＝95.047 Y n ＝100.000 Z n ＝108.883 Use the following formula to convert: The function f(t) is defined as follows: In the above formula, R′, G′, B′ represent the normalization of the R, G, B values ​​of the image, L, a, b represent the corresponding values ​​after the linear conversion of X, Y, Z, f(t) represents the conversion function, and t represents the parameter in the conversion function.

6. A textile color migration detection system according to claim 5, characterized in that: The color feature extraction module extracts the color histogram Yzft of the textile image according to the converted Lab color space, and the calculation formula is as follows: In the formula, Yzft L Represents the color histogram index of the textile image of the L channel. L represents the L channel value of the Lab color space. The range is [0, 100]. Max L Indicates the maximum L value of the color channel, Yzft a The color histogram index of the textile image representing the a channel, Yzft b Represents the color histogram index of the textile image of the b channel, a represents the a channel value of the Lab color space, the range is [-128, 127], b represents the b channel value of the Lab color space, the range is [-128, 127], n L 、n a 、n b They respectively represent the number of intervals of channel L, the number of intervals of channel a, and the number of intervals of channel b, which are set by the system.

7. A textile color migration detection system according to claim 6, characterized in that: The color feature extraction module extracts the color features of the textile image according to the converted Lab color space and calculates the Euclidean distance Ojjl of the color. The calculation formula is as follows: In the formula, Ojjl represents the Euclidean distance of colors, L1, a1, b1 represent the Lab value of the first color, and L2, a2, b2 represent the Lab value of the second color.

8. A textile color migration detection system according to claim 7, characterized in that: The color feature extraction module extracts the color features of the textile image according to the converted Lab color space and calculates the standard value ΔE of the difference between colors. The calculation formula is as follows: In the formula, ΔE represents the standard value of the difference between colors, and ΔL, Δa, and Δb represent the differences in brightness, red-green components, and blue-yellow components between two pixels in the Lab color space.

9. A textile color migration detection system according to claim 8, characterized in that: The migration detection module sets a color migration threshold Qyyz, which is used to compare with the standard value ΔE of the difference between colors to determine whether the color of the textile has migrated. The comparison method is as follows: When the standard value ΔE of the difference between colors exceeds the color migration threshold Qyyz set by the migration detection module, it indicates that the color of the textile has migrated.

10. A method for detecting color migration of textiles, characterized in that: The following steps are involved: S1, the image acquisition module obtains the textile image data Fztx through a high-precision scanner, and transmits the acquired data to the image processing module through the network; S2, the image processing module performs noise removal, brightness adjustment and contrast enhancement operations on the textile image data Fztx transmitted by the image acquisition module in sequence, to obtain a denoised textile image Qzfz, a brightness adjusted textile image Ldtz and a contrast enhanced textile image Dbzq; S3, the color space conversion module converts the image from RGB color space to Lab color space through an algorithm formula based on the contrast enhanced textile image Dbzq, which is used to analyze the color change of textiles; S4, the color feature extraction module extracts the color features of the textile image according to the converted Lab color space, including the color histogram Yzft and the Euclidean distance Ojjl of the color, and calculates the standard value ΔE of the difference between the colors and transmits the value to the migration detection module; S5, the migration detection module sets a color migration threshold Qyyz for comparison with the standard value ΔE of the color difference to determine whether the color of the textile has migrated, and generates a color migration detection report to be transmitted to the visualization module; S6. The visualization module displays the color migration detection report on the visualization terminal.

Images