Figure data analysis-based fiber identification system in antibacterial polyester-cotton composite fiber

Abstract

The invention discloses a system for identifying fibers in antibacterial polyester-cotton composite fibers based on graphic data analysis, and the system comprises a composite fiber to-be-detected sample terminal, a dual-channel microscopic imaging terminal, a fiber image preprocessing and segmenting terminal, and a fiber classification and identification terminal. The problem that fiber morphology and texture features cannot be comprehensively obtained in a single imaging mode is solved, the weight of a regularization item is dynamically adjusted by using an adaptive distance regularization level set algorithm, and the problem that a traditional segmentation method is inaccurate in segmentation of a fiber adhesion region is solved. Meanwhile, accurate division and background separation of a fiber region are realized through a deep learning network model and an OTSU adaptive threshold algorithm, and finally, classification recognition training output is performed through multi-feature fusion recombination and by using a support vector machine, so that the problem of low recognition refinement degree in composite fiber recognition is solved, and the recognition precision of the composite fiber is improved. And the recognition precision of the system on the composite fiber is obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of textile fabric identification, and in particular to a fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis. Background Art

[0002] Fiber identification in antibacterial polyester-cotton composite fibers refers to the process of accurately distinguishing and identifying different fiber components in composite fiber materials composed of polyester (polyester fiber) and cotton fibers, which have been specially treated to have antibacterial functions. This process usually involves a comprehensive analysis and judgment of the chemical composition, physical structure, performance characteristics, and antibacterial effects of the fibers. Currently, image recognition technology is mostly used for the identification of fibers and fabric tissues in textiles and textiles. By collecting images of antibacterial polyester-cotton composite fibers and using image processing algorithms and machine learning models to analyze and process the images, automatic fiber identification can be achieved.

[0003] However, in composite fibers, it is often necessary to identify multiple fiber components at the same time. These different fiber components have different characteristics and backgrounds. Most composite fabrics use certain adhesives during the processing, and the adhesives are difficult to completely remove during the fiber component quantification process, which will lead to unavoidable multi-fiber crossing and adhesion problems in the fiber identification process, causing some fiber features to be obscured. Most of the time, it is necessary to identify and classify fibers in the absence of features, and each fiber component has unique physical, chemical and optical properties. This not only requires image recognition technology to have high accuracy and resolution, but also needs to be able to distinguish and understand the subtle differences between these different fiber components and how they interact in composite structures. The existing fiber recognition system has a low degree of recognition refinement when identifying image data obtained in complex situations such as multi-fiber crossing and covering, resulting in poor accuracy in fiber recognition in multiple fibers, and ultimately low accuracy in correct classification. Summary of the Invention

[0004] The purpose of the present invention is to provide a fiber identification system in antibacterial polyester-cotton composite fibers based on graphic data analysis to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a fiber identification system in antibacterial polyester-cotton composite fibers based on graphic data analysis, comprising the following terminals: a composite fiber sample terminal to be tested, a dual-channel microscopic imaging terminal, a fiber image preprocessing and segmentation terminal, and a fiber classification and identification terminal; the composite fiber sample terminal to be tested is used to store the original composite fiber sample to be tested; the dual-channel microscopic imaging terminal is used to collaboratively collect composite fiber image feature information through multimodal optical signals; the fiber image preprocessing and segmentation terminal is used to retrieve the target composite fiber image data set from the composite fiber sample imaging database for region segmentation and corresponding labeling; the fiber classification and identification terminal is used to fuse and reorganize the morphological characteristics and texture characteristics of the composite fiber, and use a support vector machine to perform fiber classification and identification.

[0006] Preferably, the specific operation process of the dual-channel microscopic imaging terminal includes the following steps: Step S11: inputting the original composite fiber sample to be tested into the confocal microscope in the first channel, scanning the sample layer by layer with a laser, constructing a three-dimensional fiber structure based on the three-dimensional fiber volume data generated by the scanning, and obtaining a tomographic image of the sample to be tested; Step S12: rotating the polarizer group based on the polarized light imaging unit, sequentially performing acquisition triggering on the tomographic image of the sample to be tested at four polarization angles of 0°, 45°, 90°, and 135°, to obtain the maximum contrast fiber image of the sample to be tested at different angles; Step S13: synchronously inputting the original composite fiber sample to be tested into the second channel, acquiring a scattering spectrum image of the sample to be tested based on a confocal microscope, and using a multispectral fluorescence labeling unit to excite the fluorescent marker in the sample to be tested to obtain fluorescence spectrum imaging of the sample to be tested at different wavelengths; Step S14: transmitting the maximum contrast fiber image of the sample to be tested obtained by the first channel to the host computer module 1, and transmitting the scattering spectrum image and fluorescence spectrum imaging of the sample to be tested obtained by the second channel to the host computer module 2; Step S15: Integrate and summarize the image data of the sample to be tested from the host computer module 1 and the host computer module 2 to obtain a composite fiber image data set, and store it in a composite fiber sample to be tested imaging database.

[0007] Preferably, the operation process of the fiber image preprocessing and segmentation terminal includes the following steps: Step S21: dividing the target composite fiber image into connected or disconnected fiber regions based on the composite fiber target region division module, and dividing the isolated fiber region and the fiber adhesion region using the constructed deep learning network model; Step S22: Initializing the level set function using the adaptive distance regularization level set algorithm module, setting its zero level set as the initial contour, selecting the target image based on the labeled multi-region binary image, calculating the image gradient, and calculating the weight of the adaptive distance regularization term based on the gradient information, iteratively updating the level set function based on the gradient information and the adaptive distance regularization term until convergence, and the zero level set obtained after the level set function converges is the final segmentation result; Step S23: Direct indicators in the composite fiber morphology feature measurement and extraction module include: fiber diameter, scale height, scale perimeter, and scale area; relative indicators include: fiber diameter-to-height ratio, scale density, scale relative perimeter, and scale relative area. Direct indicators are directly measured and recorded, and relative indicators are calculated using the linear and nonlinear relationships between direct indicators and relative indicators. Step S24: the texture feature extraction and calculation module uses the labeled multi-region binary image to delineate the region of interest for extracting texture features, calculates the gray level co-occurrence matrix of the region of interest and feature parameters based on the gray level co-occurrence matrix, and quantifies the texture feature parameters.

[0008] Preferably, the operation process of the fiber classification and identification terminal includes the following steps: Step S31: The multi-feature fusion module normalizes the extracted morphological features and texture features of the composite fiber, defining the morphological features and texture features within a unified range. The feature array reorganization module then fuses the features into a higher-dimensional feature array; that is, the morphological features and texture features of the composite fiber are normalized to a range between 0 and 1, and reorganized into a new dimensionless five-dimensional feature array. Step S32: After the morphological features and texture features of the composite fiber are normalized, each n group of new five-dimensional feature arrays composed of the normalized morphological features and texture features is processed by a five-fold cross validation method; Step S33: In the support vector machine training model, the Gaussian kernel radial basis function is used to automatically determine the weight of each feature vector, and the feature array processed by the five-fold cross-validation method is sequentially input into the support vector machine for training and recognition, and the final classification and recognition result is output.

[0009] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention realizes the collaborative acquisition of multimodal optical signals through a dual-channel microscopic imaging terminal, solves the problem that a single imaging method cannot fully obtain the fiber morphology and texture characteristics, and combines three-dimensional structure reconstruction, polarized light imaging and fluorescence spectrum imaging to improve the resolution and information richness of the composite fiber image. The texture feature parameters are calculated by the grayscale co-occurrence matrix, which solves the problem of insufficient quantification of texture features by traditional methods. The morphological features and texture features are normalized and reorganized into a high-dimensional feature array through a multi-feature fusion module, which solves the problem of low classification accuracy of a single feature. Finally, a support vector machine is used in combination with a Gaussian kernel radial basis function to automatically determine the feature weights, and a five-fold cross-validation method is used to optimize model training to ensure the reliability and generalization ability of the classification results, thereby improving the accuracy and robustness of classification recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 Schematic diagram of the system module composition provided by an embodiment of the present invention; Figure 2 Schematic diagram of the operation flow of the dual-channel microscopic imaging terminal provided by an embodiment of the present invention; Figure 3 Schematic diagram of the operation flow of the fiber image preprocessing and segmentation terminal provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of the operation flow of the fiber classification and identification terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0012] The embodiment of the present invention is combined with Figure 1 As shown, the following technical solutions are provided: a fiber identification system in antibacterial polyester-cotton composite fibers based on graphic data analysis, including the following terminals: a composite fiber sample terminal to be tested, a dual-channel microscopic imaging terminal, a fiber image preprocessing and segmentation terminal, and a fiber classification and identification terminal.

[0013] In this embodiment, the composite fiber test sample terminal is used to store the original composite fiber test sample; the dual-channel microscopic imaging terminal is used to collaboratively collect composite fiber image feature information through multimodal optical signals: that is, the core function of the dual-channel microscopic imaging terminal is to simultaneously capture the physical structure morphology and chemical and optical properties of each layer of fiber in the composite fiber, providing differentiated feature input for subsequent algorithms, and further eliminating background noise in the single fiber imaging mode through complementary design at the optical hardware level, thereby solving the problems of missing feature information and blurred material distinction in the scenario of cross-coverage of composite fibers.

[0014] Exemplarily, a dual-channel microscopic imaging terminal includes a first channel and a second channel, wherein the first channel includes a confocal microscope, a polarized light imaging unit, and a host computer module, wherein the confocal microscope is used to reconstruct the three-dimensional structure of the fiber by using a laser to scan layer by layer and filter out the stray light on the out-of-focus plane using a pinhole to obtain a tomographic image, and the polarized light imaging unit is used to insert a rotatable linear polarizer into the confocal light path and obtain a maximum contrast fiber image by rotating the polarizer group, specifically triggering the fiber image acquisition in sequence based on the configured four-way polarization angles including 0°, 45°, 90°, and 135°, and the host computer module is used to transmit the first channel through a data transmission device. The fiber imaging data is displayed in real time on the computer; the second channel includes a Raman spectrum point scanning unit, a multi-spectral fluorescence labeling unit, and a host computer module 2, wherein the Raman spectrum point scanning unit is used to eliminate stray light in the defocused area of the original fiber sample to be tested through a confocal microscope, and obtain the structure, composition, and state information of the original fiber sample to be tested based on the generated scattering spectrum, and provide a high-resolution spectrum diagram of the original fiber sample to be tested. The multi-spectral fluorescence labeling unit is used to simultaneously obtain fluorescence spectrum imaging information of the original fiber sample to be tested at different wavelengths. The host computer module 2 is used to display the fiber imaging data of the second channel in real time on the computer through a data transmission device.

[0015] For example, combined Figure 2 As shown, the specific operation process of the dual-channel microscopic imaging terminal includes the following steps: Step S11: inputting the original composite fiber sample to be tested into the confocal microscope in the first channel, scanning the sample layer by layer with a laser, constructing the three-dimensional fiber structure based on the three-dimensional fiber volume data generated by the scanning, and obtaining a tomographic image of the sample to be tested; Step S12: rotating the polarizer group based on the polarized light imaging unit, sequentially triggering the acquisition of the tomographic image of the sample to be tested at four polarization angles of 0°, 45°, 90°, and 135°, and obtaining the maximum contrast fiber image of the sample to be tested at different angles; Step S13: synchronously inputting the original composite fiber sample to be tested into the second channel, acquiring a scattering spectrum image of the sample to be tested based on a confocal microscope, and using a multispectral fluorescence labeling unit to excite the fluorescent marker in the sample to be tested to obtain fluorescence spectrum imaging of the sample to be tested at different wavelengths; Step S14: transmitting the maximum contrast fiber image of the sample to be tested obtained by the first channel to the host computer module 1, and transmitting the scattering spectrum image and fluorescence spectrum imaging of the sample to be tested obtained by the second channel to the host computer module 2; Step S15: Integrate and summarize the image data of the sample to be tested from the host computer module 1 and the host computer module 2 to obtain a composite fiber image data set, and store it in a composite fiber sample to be tested imaging database.

[0016] In this embodiment, the fiber image preprocessing and segmentation terminal is connected to the dual-channel microscopic imaging terminal network, and the fiber image preprocessing and segmentation terminal is used to retrieve the target composite fiber image dataset from the composite fiber sample imaging database for region segmentation and corresponding labeling; Exemplarily, the fiber image preprocessing and segmentation terminal includes: a composite fiber target area division module, an adaptive distance regularized level set algorithm module, a composite fiber morphological feature measurement and extraction module, and a texture feature extraction and calculation module, wherein: the composite fiber target area division module is used to divide and mark the target composite fiber image into multiple connected areas; the adaptive distance regularized level set algorithm module is used to dynamically adjust the weight of the regularization term using image gradient information to obtain a binary image of the fiber contour of each target area; the composite fiber morphological feature measurement and extraction module is used to measure and extract the morphological features of the composite fiber including direct indicators and relative indicators; the texture feature extraction and calculation module is used to extract the grayscale co-occurrence matrix and calculate the feature parameters based on the grayscale co-occurrence matrix.

[0017] For example, combined Figure 3 As shown, the operation process of the fiber image preprocessing and segmentation terminal includes the following steps: Step S21: dividing the target composite fiber image into connected or disconnected fiber regions based on the composite fiber target region division module, and further dividing the isolated fiber region and the fiber adhesion region using the constructed deep learning network model; Exemplarily, by initially defining that all pixels in an area are adjacent and have the same or similar attributes, when a pixel cannot reach another pixel through the pixels in the area, the two pixels are not in the same connected area; when a pixel can reach another pixel through the pixels in the area, the two pixels are in the same connected area, and further judging the attributes of pixels in the same connected area or not in the same connected area, when the attributes of pixels in the same connected area are judged to be the same or similar, the area is an isolated fiber area in the connected area; conversely, when the attributes of pixels not in the same connected area are judged to be the same or similar, the area is an isolated fiber area in the non-connected area; and when the attributes of pixels in the same connected area are judged to be different or dissimilar, the area is a fiber adhesion area in the connected area; when the attributes of pixels not in the same connected area are judged to be different or dissimilar, the area is a fiber adhesion area in the non-connected area.

[0018] Specifically, a deep learning network model based on the segmentation of the target composite fiber image is constructed, and the target composite fiber image to be segmented and the label data set corresponding to each fiber category initially defined in the target fiber image are input. The label data set includes a label data set with multiple labels added to multiple fiber samples in the same composite fiber image. The target composite fiber image to be segmented is preprocessed for image enhancement and denoising, and the OTSU adaptive threshold algorithm is used to generate a binary mask to preliminarily separate the image background area and the fiber area; further, each independent fiber cluster including the isolated fiber area and the adhesion fiber area is marked through 8-neighborhood connected domain analysis to form a marking matrix, and the geometric center coordinates, area and bounding box data value of each marked area are output based on the formed marking matrix to obtain a marked multi-region binary image.

[0019] Step S22: Initialize the level set function using the adaptive distance regularized level set algorithm module, set its zero level set as the initial contour, select the target image based on the labeled multi-region binary image, calculate the image gradient, and calculate the weight of the adaptive distance regularization term based on the gradient information. Iteratively update the level set function based on the gradient information and the adaptive distance regularization term until convergence. The zero level set obtained after the level set function converges is the final segmentation result.

[0020] Step S23: The direct indicators in the composite fiber morphological feature measurement and extraction module include: fiber diameter, scale height, scale perimeter, and scale area; the relative indicators include: fiber diameter-to-height ratio, scale density, scale relative perimeter, and scale relative area. Direct indicators are directly measured and recorded, and relative indicators are calculated using the linear and nonlinear relationships between direct indicators and relative indicators.

[0021] Step S24: The texture feature extraction and calculation module uses the labeled multi-region binary image to delineate a region of interest for extracting texture features, calculates a gray level co-occurrence matrix of the region of interest and feature parameters based on the gray level co-occurrence matrix, and quantifies the texture feature parameters; Exemplarily, the region of interest for extracting texture features is delineated as follows: by traversing the marked regional binary image, the circumscribed moment of the target fiber region is extracted in the grayscale image, and this is used as the region of interest. The grayscale co-occurrence matrix in the four directions of 0°, 45°, 90°, and 135° is extracted respectively, and the four obtained grayscale co-occurrence matrices are summed to obtain a new grayscale co-occurrence matrix, which is normalized. The new co-occurrence matrix extracted by summing the grayscale co-occurrence matrices reflects the texture features in each direction, and the texture feature parameters of energy, entropy, contrast, and correlation of the four normalized co-occurrence matrices are calculated. The average value is calculated for the same feature parameter, and the texture features of the fiber are described based on the average value.

[0022] In this embodiment, the fiber classification and recognition terminal is network-connected to the fiber image preprocessing and segmentation terminal. The fiber classification and recognition terminal is used to define the morphological characteristics and texture characteristics of the composite fiber within a unified range, fuse the morphological characteristics with the texture characteristics to form a higher-dimensional feature array, and identify the fiber category in the composite fiber through a support vector machine with the same configuration.

[0023] Exemplarily, the fiber classification and recognition terminal includes: a multi-feature fusion module, a feature array reorganization module, and a support vector machine training model; the multi-feature fusion module is used to fuse the morphological features and texture features of the composite fiber; the feature array reorganization module is used to perform a higher-dimensional combination of the fused feature array; the support vector machine training model is used to perform classification and recognition based on the constructed support vector machine training model, and output the final classification and recognition results.

[0024] For example, combined Figure 4 As shown, the operation process of the fiber classification and identification terminal includes the following steps: Step S31: The multi-feature fusion module normalizes the extracted morphological and textural features of the composite fiber, defining them within a unified range. The feature array reorganization module then fuses them into a higher-dimensional feature array; that is, the morphological and textural features of the composite fiber are normalized to a range between 0 and 1, and reorganized into a new dimensionless five-dimensional feature array. Step S32: After the morphological features and texture features of the composite fiber are normalized, each n group of new five-dimensional feature arrays composed of the normalized morphological features and texture features is processed by a five-fold cross validation method; Exemplarily, the specific processing method of the five-fold cross-validation method includes: processing each n groups of new five-dimensional feature arrays composed of normalized morphological features and texture features through the five-fold cross-validation method, that is, performing a five-fold cross-validation experiment on the new five-dimensional feature array, each time the five-fold cross-validation experiment is performed, all feature data samples in the five-dimensional feature array are randomly divided into five parts, each part contains the same number of morphological feature arrays and texture feature arrays, four of them are used as training data sets and one is used as a test data set for testing in turn, the recognition rate of the five tests is output, the average value of the five experimental results is calculated, the calculated average value is used as the recognition accuracy of this five-fold cross-validation, the average value of the recognition accuracy of the five five-fold cross-validation is calculated, and the average value of the recognition accuracy is used as the final recognition accuracy of the support vector machine training model.

[0025] Step S33: In the support vector machine training model, the Gaussian kernel radial basis function is used to automatically determine the weight of each feature vector, and the feature array processed by the five-fold cross-validation method is sequentially input into the support vector machine for training and recognition, and the final classification and recognition results are output.

[0026] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0027] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A fiber identification system for antibacterial polyester-cotton composite fibers based on graphical data analysis, characterized by: It includes the following terminals: a composite fiber sample terminal to be tested, a dual-channel microscopic imaging terminal, a fiber image preprocessing and segmentation terminal, and a fiber classification and identification terminal; the composite fiber sample terminal to be tested is used to store the original composite fiber sample to be tested; the dual-channel microscopic imaging terminal is used to collaboratively collect composite fiber image feature information through multimodal optical signals; the fiber image preprocessing and segmentation terminal is used to retrieve the target composite fiber image data set from the composite fiber sample imaging database for region segmentation and corresponding labeling; the fiber classification and identification terminal is used to fuse and reorganize the morphological features and texture features of the composite fiber, and use a support vector machine to perform fiber classification and identification, wherein the composite fiber sample terminal to be tested is network-connected to the dual-channel microscopic imaging terminal, the fiber image preprocessing and segmentation terminal is network-connected to the dual-channel microscopic imaging terminal, and the fiber classification and identification terminal is network-connected to the fiber image preprocessing and segmentation terminal.

2. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 1, characterized in that: The dual-channel microscopic imaging terminal includes a first channel and a second channel. The first channel includes a confocal microscope, a polarized light imaging unit, and a host computer module 1; the second channel includes a Raman spectrum point scanning unit, a multi-spectral fluorescence labeling unit, and a host computer module 2.

3. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 2, characterized in that: The specific operation process of the dual-channel microscopic imaging terminal includes the following steps: Step S11: inputting the original composite fiber sample to be tested into the confocal microscope in the first channel, scanning the sample layer by layer with a laser, constructing a three-dimensional fiber structure based on the three-dimensional fiber volume data generated by the scanning, and obtaining a tomographic image of the sample to be tested; Step S12: rotating the polarizer group based on the polarized light imaging unit, sequentially performing acquisition triggering on the tomographic image of the sample to be tested at four polarization angles of 0°, 45°, 90°, and 135°, to obtain the maximum contrast fiber image of the sample to be tested at different angles; Step S13: synchronously inputting the original composite fiber sample to be tested into the second channel, acquiring a scattering spectrum image of the sample to be tested based on a confocal microscope, and using a multispectral fluorescence labeling unit to excite the fluorescent marker in the sample to be tested to obtain fluorescence spectrum imaging of the sample to be tested at different wavelengths; Step S14: transmitting the maximum contrast fiber image of the sample to be tested obtained by the first channel to the host computer module 1, and transmitting the scattering spectrum image and fluorescence spectrum imaging of the sample to be tested obtained by the second channel to the host computer module 2; Step S15: Integrate and summarize the image data of the sample to be tested from the host computer module 1 and the host computer module 2 to obtain a composite fiber image data set, and store it in a composite fiber sample to be tested imaging database.

4. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 3, characterized in that: The fiber image preprocessing and segmentation terminal includes: a composite fiber target area division module, an adaptive distance regularized level set algorithm module, a composite fiber morphological feature measurement and extraction module, and a texture feature extraction and calculation module; The composite fiber target area division module is used to divide and mark the target composite fiber image into multiple connected areas; the adaptive distance regularization level set algorithm module is used to dynamically adjust the weight of the regularization term using image gradient information to obtain a binary image of the fiber contour of each target area; the composite fiber morphological feature measurement and extraction module is used to measure and extract the morphological features of the composite fiber, including direct indicators and relative indicators; the texture feature extraction and calculation module is used to extract the grayscale co-occurrence matrix and calculate the feature parameters based on the grayscale co-occurrence matrix.

5. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 4, characterized in that: The operation process of the fiber image preprocessing and segmentation terminal includes the following steps: Step S21: dividing the target composite fiber image into connected or disconnected fiber regions based on the composite fiber target region division module, and dividing the isolated fiber region and the fiber adhesion region using the constructed deep learning network model; Step S22: Initializing the level set function using the adaptive distance regularization level set algorithm module, setting its zero level set as the initial contour, selecting the target image based on the labeled multi-region binary image, calculating the image gradient, and calculating the weight of the adaptive distance regularization term based on the gradient information, iteratively updating the level set function based on the gradient information and the adaptive distance regularization term until convergence, and the zero level set obtained after the level set function converges is the final segmentation result; Step S23: Direct indicators in the composite fiber morphology feature measurement and extraction module include: fiber diameter, scale height, scale perimeter, and scale area; relative indicators include: fiber diameter-to-height ratio, scale density, scale relative perimeter, and scale relative area. Direct indicators are directly measured and recorded, and relative indicators are calculated using the linear and nonlinear relationships between direct indicators and relative indicators. Step S24: the texture feature extraction and calculation module uses the labeled multi-region binary image to delineate the region of interest for extracting texture features, calculates the gray level co-occurrence matrix of the region of interest and feature parameters based on the gray level co-occurrence matrix, and quantifies the texture feature parameters.

6. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 5, characterized in that: The method of using the constructed deep learning network model to divide the isolated fiber area and the fiber adhesion area includes: constructing a deep learning network model based on the target composite fiber image division, inputting the target composite fiber image to be divided and the label data set corresponding to each fiber category initially defined in the target fiber image, the label data set including a label data set with multiple labels added to multiple fiber samples in the same composite fiber image, performing image enhancement and denoising preprocessing on the target composite fiber image to be divided, and using the OTSU adaptive threshold algorithm to generate a binary mask to preliminarily separate the image background area and the fiber area; further marking each independent fiber cluster including the isolated fiber area and the adhesion fiber area through 8-neighborhood connected domain analysis to form a label matrix, and outputting the geometric center coordinates, area and bounding box data value of each marked area based on the formed label matrix to obtain a labeled multi-region binary image.

7. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 6, characterized in that: The method for demarcating the region of interest for extracting texture features includes: traversing the marked regional binary image, extracting the circumscribed moment of the target fiber region in the grayscale image, taking the circumscribed moment as the region of interest, extracting the grayscale co-occurrence matrix in four directions of 0°, 45°, 90°, and 135°, respectively, summing the four obtained grayscale co-occurrence matrices to obtain a new grayscale co-occurrence matrix, and normalizing it. The new co-occurrence matrix extracted by summing the four grayscale co-occurrence matrices reflects the texture features in each direction, calculates the texture feature parameters of energy, entropy, contrast, and correlation of the four normalized co-occurrence matrices, calculates the average value for the same feature parameter, and describes the texture features of the fiber based on the average value.

8. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 7, characterized in that: The fiber classification and identification terminal includes: a multi-feature fusion module, a feature array reorganization module, and a support vector machine training model; the multi-feature fusion module is used to fuse the morphological features and texture features of the composite fiber; the feature array reorganization module is used to perform a higher-dimensional combination of the fused feature arrays; the support vector machine training model is used to perform classification and identification based on the constructed support vector machine training model, and output the final classification and identification results.

9. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 8, characterized in that: The operation process of the fiber classification and identification terminal includes the following steps: Step S31: The multi-feature fusion module normalizes the extracted morphological features and texture features of the composite fiber, defining the morphological features and texture features within a unified range. The feature array reorganization module then fuses the features into a higher-dimensional feature array; that is, the morphological features and texture features of the composite fiber are normalized to a range between 0 and 1, and reorganized into a new dimensionless five-dimensional feature array. Step S32: After the morphological features and texture features of the composite fiber are normalized, each n group of new five-dimensional feature arrays composed of the normalized morphological features and texture features is processed by a five-fold cross validation method; Step S33: In the support vector machine training model, the Gaussian kernel radial basis function is used to automatically determine the weight of each feature vector, and the feature array processed by the five-fold cross-validation method is sequentially input into the support vector machine for training and recognition, and the final classification and recognition result is output.

10. The fiber identification system for antibacterial polyester-cotton composite fibers based on graphic data analysis according to claim 9, characterized in that: The specific processing method of the five-fold cross-validation method includes: processing each n groups of new five-dimensional feature arrays composed of normalized morphological features and texture features through the five-fold cross-validation method, that is, performing a five-fold cross-validation experiment on the new five-dimensional feature array, each time the five-fold cross-validation experiment is performed, all feature data samples in the five-dimensional feature array are randomly divided into five parts, each part contains the same number of morphological feature arrays and texture feature arrays, four of the parts are used as training data sets and one is used as a test data set for testing in turn, the recognition rate of the five tests is output, the average value of the five experimental results is calculated, the calculated average value is used as the recognition accuracy of this five-fold cross-validation, the average value of the recognition accuracy of the five five-fold cross-validation is calculated, and the average value of the recognition accuracy is used as the final recognition accuracy of the support vector machine training model.

Images