Pineapple leaf fiber fabric evaluation method based on multispectral image

Abstract

The invention provides a pineapple leaf fiber fabric evaluation method based on a multispectral image, and relates to the technical fabric evaluation field, the pineapple leaf fiber fabric evaluation method comprises the following steps: respectively shooting image data of a pineapple fiber fabric in a black background and a white background, analyzing the image data, generating a gray scale, and analyzing the light transmission of the fabric through the gray scale difference; the method comprises the following steps of: dividing a fabric into nine areas, generating a light transmission index, analyzing gray levels, generating a mean gray level and a deviation gray level, dividing the fabric into nine areas, and performing standardized color value processing on each area, according to the method, the color deviation of the fabric in each area is analyzed in the color space, the area color difference is generated, finally, the fabric evaluation index for evaluating the fabric grade is generated through comprehensive analysis, finally, the fabric quality evaluation grade is output, manual naked eye observation is not needed, and evaluation is rapid and accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of fabric evaluation, and specifically to a method for evaluating pineapple leaf fiber fabric based on multi-spectral images. Background Art

[0002] Pineapple leaf fiber fabric, also known as "pineapple fiber", is a natural fiber material extracted from the leaves of pineapple plants. Pineapple leaves are rich in cellulose. Through specific processing techniques such as boiling, soaking, and spinning, high-quality fibers can be effectively extracted. This fabric has good air permeability and moisture absorption, making it very popular in summer clothing and household items. In addition, the pineapple leaf fiber fabric has high strength and excellent wear resistance, which can meet the needs of daily wearing and use. Its natural properties give it an advantage in environmental protection and conform to the concept of sustainable development. Pineapple leaf fiber also has excellent antibacterial properties and can effectively inhibit the growth of bacteria. When processing pineapple leaf fiber fabric, due to uneven coating and dyeing during the processing process, the color purity and knitting uniformity of the fabric are not unified after production.

[0003] Therefore, after the pineapple leaf fiber fabric is made, it is necessary to visually inspect the color and density of the fabric by human eyes. However, the error evaluation standard based on visual inspection is relatively broad and prone to omissions, resulting in the inability to quickly and accurately evaluate the quality grade of the fabric.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for evaluating pineapple leaf fiber fabric based on multi-spectral images to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A method for evaluating pineapple leaf fiber fabric based on multi-spectral images, the specific steps include: S1. Under black and white backgrounds, the flat and unfolded fabric is photographed by a camera respectively to obtain image data. The image data is processed by Canva software to remove the non-fabric background, and the gray values of the pixel points in the image data after removing the background are obtained; S2. Perform correlation analysis on the gray values to generate gray parameters. The gray parameters include average gray and deviation gray. The average gray is used to reflect the average gray value of the image data, and the deviation gray is used to reflect the gray deviation degree of the image data; S3. Divide the fabric evenly into nine regions, conduct a correlation analysis on each region of the fabric to generate a color mean value, which is used to reflect the RGB three-channel mean values of each region of the fabric, and conduct a correlation analysis on the color mean value to generate a color standard value; S4. Conduct a correlation analysis on the color standard value to generate a regional color difference, conduct a correlation analysis on the regional color difference and the gray-scale parameter under a black background to generate a fabric color deviation evaluation index MPZ, and the fabric color deviation evaluation index is used to reflect the color purity of the fabric; S5. Conduct a correlation analysis on the gray-scale parameters under a black background and a white background to generate a light transmittance index, and the light transmittance index is used to reflect the light transmittance performance of the fabric; S6. Conduct a correlation analysis on the fabric color deviation evaluation index MPZ and the light transmittance index TGZ to generate a fabric evaluation index MGZ, compare the fabric evaluation index MGZ with a threshold value, and output the fabric evaluation grade.

[0007] Further, lay the fabric flat, with black and white backgrounds at the bottom, take pictures of the fabric respectively to obtain image data, use the Canva software, adopt the Candy algorithm to detect and identify the contour background in the image data, and remove the detected contour background through the Gaussian blur algorithm. Number the pixel points of the image data after removing the background. The total number of pixel points is N. Conduct a correlation analysis on the pixel points of the image data to generate gray scale; ; Among them, is the gray-scale value of the i-th pixel point of the image data, is the red component of the i-th pixel point of the image data, is the green component of the i-th pixel point of the image data, is the blue component of the i-th pixel point of the image data.

[0008] Further, conduct a correlation analysis on the gray-scale values to generate an average gray scale and a deviation gray scale , and the formulas are as follows: ; The average gray scale is used to reflect the average gray-scale value of the fabric image data, and the deviation gray scale is used to reflect the deviation degree of the overall gray scale of the fabric image data, that is, to reflect whether the overall brightness of the fabric is uniform.

[0009] Further, the color mean values include the red mean value R, the green mean value G, and the blue mean value B. Among them, the red mean value R is the mean value of the red components of all pixel points in the current area, the green mean value G is the mean value of the green components of all pixel points in the current area, and the blue mean value B is the mean value of the blue components of all pixel points in the current area. Correlation analysis is performed on the color mean values to generate color standard values, and the color standard values include X, Y, and Z. The formula used is: ; Among them, , , , , , , , , are the color values after normalization, , , are the color values after gamma correction, and X, Y, and Z are the color values after standardization, for unified color output.

[0010] Further, correlation analysis is performed on the color standard values to generate regional chromaticity values, and the regional chromaticity values include lightness L, red-green chroma a, and yellow-blue chroma b. The formula used is: ; Among them, , , the lightness L is used to reflect the vividness of the color in this area, the red-green chroma a is used to reflect the degree of red-green color in this area, and the yellow-blue chroma b is used to reflect the degree of yellow-blue color in this area; Correlation analysis is performed on the regional chromaticity values to generate regional color differences , and the formula used is: ; Among them, , , are the lightness, red-green chroma, and yellow-blue chroma of the j-th fabric area in the fabric, and j is used to index the area, are the average values of the lightness, red-green chroma, and yellow-blue chroma of the nine areas respectively, and the regional color difference is used to reflect the deviation degree of the color between different areas of the fabric; Correlation analysis is performed on the regional color difference and gray-scale parameters under a black background to generate a fabric color deviation evaluation index MPZ. The formula used is: ; Among them, the fabric color deviation evaluation index MPZ is used to reflect the overall color deviation degree of the fabric.

[0011] Further, the average gray scale under a black background , the deviation gray scale and the average gray scale under a white background , the deviation gray scale are subjected to correlation analysis to generate a light transmittance index TGZ. The formula is as follows: ; The light transmittance index TGZ is used to reflect the light transmittance of the fabric.

[0012] Further, a correlation analysis is performed on the fabric color deviation evaluation index MPZ and the light transmittance index TGZ to generate a fabric evaluation index MGZ. The formula is as follows: ; The fabric evaluation index MGZ is used to evaluate the fabric grade. The fabric evaluation index MGZ is compared with the threshold . When , the quality evaluation grade of the fabric is grade two. At this time, the color purity of the fabric is poor, the density is low, the light transmittance is high, and there is color deviation or dirt on it; when , the quality evaluation grade of the fabric is grade one. At this time, the color purity of the fabric is high, the density is high, and the light transmittance is low.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: By respectively taking image data of the pineapple fiber fabric under a black background and a white background, analyzing the image data to generate gray scales, analyzing the light transmittance of the fabric through the gray scale differences, generating a light transmittance index, and analyzing the gray scales to generate an average gray scale and a deviation gray scale, which reflects the overall brightness deviation degree of the fabric and indirectly reflects the flatness of the fabric. At the same time, the fabric is divided into nine regions, and the standardized color values of each region are processed. The color deviation of the fabric in each region is analyzed in the color space to generate a regional color difference. Finally, through comprehensive analysis, a fabric evaluation index for evaluating the fabric grade is generated, and finally the fabric quality evaluation grade is output, without manual visual observation, and the evaluation is fast and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0016] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0017] Example: See also Figure 1 The pineapple leaf fiber fabric of the present invention is made based on the following method: 1. Sizing: The sizing agent formula is: starch, polyvinyl alcohol (PVA), acrylate, the mass ratio is 6:3:1, the sizing concentration is 8-10%, the sizing temperature is 90-95℃, and the moisture regain of the yarn after sizing is controlled at 6-8%.

[0018] 2. Warping: Arrange the yarn into warp sheets according to the process requirements and wind them onto the warp beam. Warping speed: 300-400m / min, warp tension: 15-20cN / root, warp beam winding density: 0.5-0.6g / cm³, to ensure that the warp yarns are evenly arranged and have consistent tension, to avoid problems such as twisted ends and broken ends.

[0019] 3. Thread the warp yarn into the heald and reed according to the requirements of the plain weave. Heald density: 10-12 pieces / cm, reed number: 60-80, threading method: straight threading, and careful inspection is required when threading the heald and reed to avoid missing or wrong threading.

[0020] 4. Weaving, loom selection: Use rapier loom or air jet loom, preferably rapier loom to adapt to the characteristics of pineapple leaf fiber. Loom speed: 400-500r / min, warp tension: 20-25 cN / root, weft tension: 15-20 cN / root, opening time: 300°-320°, weft insertion time: 80°-100°, beating force: 200-250N, warp density: 50-60 roots / cm, weft density: 40-50 roots / cm, fabric structure: 1 / 1 plain weave. Control the tension of warp and weft during weaving to avoid problems such as breakage and weft shrinkage.

[0021] 5. Post-treatment, desizing: Desize the woven fabric to remove the sizing agent. Use amylase for desizing at a temperature of 60 - 70°C for 30 - 40 minutes. Dyeing: Use reactive dyes for dyeing at a temperature of 60°C for 30 minutes. Wash the fabric after dyeing to remove the floating color. Stentering: Use a stentering machine to perform heat setting on the fabric at a temperature of 150 - 160°C for 30 - 40 seconds. Use a cationic modified silicone softener for softening treatment to improve the hand feeling and comfort of the fabric.

[0022] The dyed color of the pineapple leaf fiber fabric of the present invention is a solid color.

[0023] The present invention provides a technical solution: Based on the multi-spectral image pineapple leaf fiber fabric evaluation method, data is obtained by taking multi-spectral images and the data is averaged to reduce the error value in single-spectrum shooting. The specific steps include: Step 1: Under black and white backgrounds, take pictures of the flat fabric through a multi-spectral camera at different spectra to obtain the averaged image data. The image data is processed by Canva software to remove the non-fabric background, and the gray value of the pixel points in the image data after removing the background is obtained; the resolution of the camera is 1920*1080; Lay the fabric flat, with black and white backgrounds at the bottom, take pictures of the fabric respectively to obtain image data. Use the Candy algorithm in Canva software to detect and identify the contour background in the image data, and use the Gaussian blur algorithm to remove the detected contour background. Number the pixel points in the image data after removing the background. The total number of pixel points is N. Perform correlation analysis on the pixel points of the image data to generate gray levels. ; Among them, is the gray value of the i-th pixel point of the image data, is the red component of the i-th pixel point of the image data, is the green component of the i-th pixel point of the image data, is the blue component of the i-th pixel point of the image data. The gray value can reflect the brightness of the pixel points in the image data, and the larger the value, the brighter.

[0024] Step 2: Perform correlation analysis on the gray values to generate gray parameters, which include average gray and deviation gray. The average gray is used to reflect the average gray value of the image data, and the deviation gray is used to reflect the degree of gray deviation of the image data; When there are differences in the colors on the image data, the corresponding gray values will have gaps. Therefore, perform correlation analysis on the gray values to generate the average gray and the deviation gray , based on the formula: ; Average grayscale Used to reflect the average grayscale value of the fabric image data, deviation grayscale Used to reflect the deviation degree of the overall grayscale of the fabric image data, that is, to reflect whether the overall brightness of the fabric is uniform, and can indirectly reflect whether the overall color of the fabric is consistent. The lower the deviation grayscale , the more consistent the color is.

[0025] Step 3: Grayscale is one aspect of the fabric. In order to evaluate the fabric from the aspect of color difference, the fabric is evenly divided into nine regions, and correlation analysis is performed on each region of the fabric to generate color means. The color means are used to reflect the RGB three-channel means of each region of the fabric. Correlation analysis is performed on the color means to generate color standard values; The color means include red mean R, green mean G, and blue mean B. Among them, the red mean R is the mean of the red components of all pixel points in the current region, the green mean G is the mean of the green components of all pixel points in the current region, and the blue mean B is the mean of the blue components of all pixel points in the current region. Correlation analysis is performed on the color means to generate color standard values. The color standard values include X, Y, and Z. First, the RGB components are normalized to generate normalized values , , , and then gamma correction is performed on the normalized values to generate , , , and then through the contribution matrix to , , for processing to generate color standard values X, Y, and Z. The formula is: ; Among them, , , , , , , , , are the normalized color values, , , is the color value after gamma correction, and X, Y, and Z are the standardized color values, which are used to unify the color output. The normalized color value facilitates data processing. Since the color space adopts a non-linear photometric response, directly using the standardized RGB values for conversion will result in inaccurate results. Therefore, gamma correction is required to convert the RGB values to a linear space. The contribution matrix is defined based on the CIE1931 color space. The generated color standard values X, Y, and Z after processing can uniformly manage colors on all devices. Among them, X represents the stimulation value of red, which is related to the wavelength of the red spectrum and its perception by the human eye. Y represents the lightness value, which usually corresponds to the human eye's perception of brightness. The higher the value, the brighter the color. Z represents the stimulation value of blue, which is related to the wavelength of the blue spectrum and its perception by the human eye.

[0026] Step 4: Conduct a correlation analysis on the color standard values to generate regional color differences. Conduct a correlation analysis on the regional color differences and gray-scale parameters under a black background to generate a fabric color deviation evaluation index MPZ. The fabric color deviation evaluation index is used to reflect the color purity of the fabric; Conduct a correlation analysis on the color standard values to generate regional chromaticity values. The regional chromaticity values include lightness L, red-green chroma a, and yellow-blue chroma b. The formulas are as follows: ; Among them, ,, The function is used to convert the relative luminance value to the perceived luminance, is used to distinguish bright and dark areas. When x is greater than this value, cube root processing is adopted; when x is not greater than this value, linear processing is adopted, where x is the independent variable. The regional chromaticity values are all used to reflect the color conditions of nine regions on the fabric. Among them, lightness L is used to reflect the vividness of the color in this region, red-green chroma a is used to reflect the red-green color degree in this region, and yellow-blue chroma b is used to reflect the yellow-blue color degree in this region; by analyzing whether there are differences in the color vividness, red-green color degree, and yellow-blue color degree between regions, it is determined whether the overall color of the fabric is color-biased. Therefore, a correlation analysis is conducted on the regional chromaticity values to generate regional color differences The formulas are as follows: ; Among them, , , The lightness, red-green chroma, and yellow-blue chroma of the j-th fabric region in the fabric. j is used to index the region, are respectively the averages of the lightness, red-green chroma, and yellow-blue chroma of the nine regions. The regional color difference is used to reflect the deviation degree of colors between regions of the fabric. The regional color difference The larger the value, the higher the color deviation degree of the nine regions of the fabric, that is, the less pure the color; Under a black background, a correlation analysis is performed on the regional color difference and gray-scale parameters under the black background to generate a fabric color deviation evaluation index MPZ. The formula is as follows: ; Among them, the fabric color deviation evaluation index MPZ is used to reflect the overall color and brightness deviation degree of the fabric. The smaller the fabric color deviation evaluation index, the smaller the deviation of the fabric color and the fabric brightness, that is, the flatter the fabric, the purer the color, and the better the quality.

[0027] Step 5: Perform a correlation analysis on the gray-scale parameters under the black background and the white background to generate a light transmittance index, and the light transmittance index is used to reflect the light transmittance performance of the fabric; Among them, in the image data, under the black background of the pineapple leaf fiber fabric, the light-transmitting part is black, and its three-color values are 0, 0, 0. Under the white background, the light-transmitting part is white, and the three-color values of white are 255, 255, 255. Based on this difference, the greater the brightness difference of the corresponding pixel points on the black image data and the white image data, the more pixel points in the light-transmitting part.

[0028] Therefore, for the average gray scale under the black background , deviation gray scale and the average gray scale under the white background , deviation gray scale perform a correlation analysis to generate a light transmittance index TGZ. The formula is as follows: ; The light transmittance index TGZ is used to reflect the light transmittance of the fabric. Under a black background, the average gray level indicates the ability of the material to absorb light. The higher the gray level, the stronger the ability of the material to reflect or transmit light. The deviation gray level reflects the degree of dispersion of the gray level values. By summing up the deviation gray levels, the stability of the gray level values under different backgrounds can be comprehensively reflected. Through the gray level difference, the light transmittance performance can be evaluated. The larger the light transmittance index, the greater the light transmittance, indicating a lower fabric density. When the difference in the average gray level between the black background and the white background is large, it shows that the overall brightness difference of the image is significant under different backgrounds, which indicates that there are obvious changes in the light transmittance of the image. The better the light transmittance index, the difference in the deviation gray level reflects the consistency of the gray level distribution of the image. If the difference in the deviation gray level is large, it means that under different backgrounds, the degree of change of the gray level values is different, which is likely to interfere with the analysis of the light transmittance under the black and white backgrounds. The greater the degree of deviation, the less accurate the light transmittance index. That is, the greater the average gray level deviation between the black and white backgrounds, the better the light transmittance, the larger the light transmittance index TGZ, the greater the deviation gray level between the black and white backgrounds, the higher the degree of interference with the light transmittance evaluation, the better and more difficult to evaluate the light transmittance, and the smaller the light transmittance index TGZ.

[0029] Step 6: Conduct a correlation analysis on the fabric color deviation evaluation index MPZ and the light transmittance index TGZ to generate the fabric evaluation index MGZ, compare the fabric evaluation index MGZ with the threshold value, and output the fabric evaluation grade.

[0030] Conduct a correlation analysis on the fabric color deviation evaluation index MPZ and the light transmittance index TGZ to generate the fabric evaluation index MGZ. The formula is as follows: ; Determine the influence degree of the light transmittance index TGZ on the fabric color deviation evaluation index through the light transmittance weight factor. When is greater than 0, the light transmittance index TGZ is positively correlated with the fabric color deviation evaluation index. The larger the light transmittance index TGZ, the larger the fabric evaluation index MGZ, and the worse the quality evaluation. The larger the fabric color deviation evaluation index MPZ, the greater the fabric color deviation, the larger the fabric evaluation index MGZ, and the worse the quality evaluation. The fabric evaluation index MGZ is used to evaluate the fabric grade. is the light transmittance weight factor, and the range is set to , which is used to evaluate the importance of the fabric light transmittance. When set to 0, it means that the light transmittance is not important for fabric evaluation. The larger the value, the higher the importance of the light transmittance participating in the quality evaluation. The threshold value is used to define the boundary of the fabric evaluation quality grade, which is obtained by analyzing the samples. Compare the fabric evaluation index MGZ with the threshold value When When the fabric quality assessment level is secondary, the color purity of the fabric is poor, the density is low, the light transmittance is high, and there is color deviation or dirt on it; when When the fabric quality assessment level is primary, the color purity of the fabric is high, the density is high, and the light transmittance is low.

[0031] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0032] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0033] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0034] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A method for evaluating pineapple leaf fiber fabrics based on multispectral images, characterized in that, The specific steps include: S1. Lay the pineapple leaf fiber fabric separately against pure black and white backgrounds, photograph the flat and unfolded fabric with a camera to obtain image data. Use Canva software to remove the non-fabric background from the image data, and obtain the gray values of the pixel points in the image data after removing the background. S2. Conduct a correlation analysis on the gray values to generate gray parameters, which include the average gray and the deviation gray. The average gray is used to reflect the average gray value of the image data, and the deviation gray is used to reflect the degree of gray deviation of the image data. S3. Against the black background, evenly divide the fabric into nine regions, conduct a correlation analysis on each region of the fabric to generate color means, which are used to reflect the RGB three-channel means of each region of the fabric. Conduct a correlation analysis on the color means to generate color standard values. S4. Conduct a correlation analysis on the color standard values to generate regional color differences. Conduct a correlation analysis on the regional color differences and the gray parameters against the black background to generate a fabric color deviation evaluation index MPZ, which is used to reflect the color purity of the fabric. S5. Conduct a correlation analysis on the gray parameters against the black background and the white background to generate a light transmittance index, which is used to reflect the light transmittance performance of the fabric. S6. Conduct a correlation analysis on the fabric color deviation evaluation index MPZ and the light transmittance index TGZ to generate a fabric evaluation index MGZ. Compare the fabric evaluation index MGZ with a threshold value and output the fabric evaluation grade.

2. The method for evaluating pineapple leaf fiber fabric based on multispectral images according to claim 1, wherein: Lay the fabric flat and spread it out, with black and white backgrounds at the bottom. Photograph the fabric separately to obtain image data. Use Canva software and the Candy algorithm to detect and identify the contour background in the image data, and use the Gaussian blur algorithm to remove the detected contour background. Number the pixel points in the image data after removing the background. The total number of pixel points is N. Conduct a correlation analysis on the pixel points of the image data to generate gray values. ； wherein, is the gray value of the i-th pixel point of the image data, is the red component of the i-th pixel point of the image data, is the green component of the i-th pixel point of the image data, is the blue component of the i-th pixel point of the image data, is the index of the pixel point in the image data after removing the background, and .

3. The method for evaluating pineapple leaf fiber fabric based on multispectral images according to claim 2, wherein: Perform a correlation analysis on the gray values to generate the average gray value and the deviation gray value , and the formula is as follows: ； Average gray scale Used to reflect the average gray scale value of the fabric image data, deviation gray scale Used to reflect the degree of deviation of the overall gray scale of the fabric image data, that is, to reflect whether the overall brightness of the fabric is uniform.

4. The pineapple leaf fiber fabric evaluation method based on multi-spectral images according to claim 1, characterized in that: The color means include the red mean R, the green mean G, and the blue mean B. Among them, the red mean R is the mean of the red components of all pixel points in the current region, the green mean G is the mean of the green components of all pixel points in the current region, and the blue mean B is the mean of the blue components of all pixel points in the current region. Conduct a correlation analysis on the color means to generate color standard values, which include X, Y, and Z. The formula is as follows: ； Among them, , , , , , , , , are the normalized color values, , , are the color values after gamma correction, and X, Y, Z are the standardized color values for unified color output.

5. The method for evaluating pineapple leaf fiber fabric based on multi-spectral images according to claim 4, characterized in that: Conduct a correlation analysis on the color standard values to generate regional chromaticity values, which include lightness L, red-green chroma a, and yellow-blue chroma b. The formula is as follows: ； Among them, , , the lightness L is used to reflect the color vividness of this area, the red-green degree a is used to reflect the red-green color degree of this area, and the yellow-blue degree b is used to reflect the yellow-blue color degree of this area; Perform a correlation analysis on the regional chromaticity values to generate regional color differences , and the formula used is as follows: ； Among them, , , are the lightness, red-green degree, and yellow-blue degree of the j-th region in the fabric, respectively. j is used to index the regions. are the averages of the lightness, red-green degree, and yellow-blue degree of the nine regions, respectively. The regional color difference is used to reflect the deviation degree of the color and luster between the regions of the fabric. Conduct a correlation analysis on the regional color differences and the gray parameters against the black background to generate a fabric color deviation evaluation index MPZ. The formula is as follows: ； Among them, the fabric color deviation evaluation index MPZ is used to reflect the overall color deviation degree of the fabric.

6. The method for evaluating pineapple leaf fiber fabric based on multi-spectral images according to claim 5, wherein: Average gray scale against a black background , deviation gray scale and average gray scale against a white background , deviation gray scale Perform a correlation analysis to generate a light transmittance index TGZ, based on the formula: ； The light transmittance index TGZ is used to reflect the light transmittance of the fabric.

7. The method for evaluating pineapple leaf fiber fabric based on multispectral images according to claim 6, wherein: Conduct a correlation analysis on the fabric color deviation evaluation index MPZ and the light transmittance index TGZ to generate a fabric evaluation index MGZ. The formula is as follows: ； The fabric evaluation index MGZ is used to evaluate the fabric grade. The fabric evaluation index MGZ is compared with the threshold value for comparison. When occurs, the quality evaluation grade of the fabric is secondary. At this time, the color purity of the fabric is poor, the density is low, the light transmittance is high, and there are color deviations or stains on it; when occurs, the quality evaluation grade of the fabric is primary. At this time, the color purity of the fabric is high, the density is high, and the light transmittance is low.

Images