Manufacturing method and equipment of dispensing adhesive fabric and medium

Abstract

The invention provides a manufacturing method, equipment and medium of a dispensing glue fabric, and relates to the technical field of dispensing glue fabric manufacturing, and the method comprises the steps: dividing a target image into a plurality of sub-images; performing DCT (Discrete Cosine Transform) on each sub-image to obtain a DCT coefficient matrix corresponding to each sub-image; according to the characteristics of each DCT coefficient matrix, all the DCT coefficient matrixes are clustered; processing each DCT coefficient matrix in the QA to obtain a processed DCT coefficient matrix cluster WA corresponding to the QA; mapping an energy center corresponding to each DCT coefficient matrix in the WA into the target image to obtain a mass center of a glue point corresponding to each DCT coefficient matrix in the QA; the mass center of a complete glue point composed of the incomplete glue points is obtained; according to the mass center distribution condition of each glue point in the target image, determining whether the to-be-detected point glue sol fabric is abnormal or not; the final fabric is obtained; according to the method, the data processing amount can be greatly reduced, and the detection efficiency is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of point sol manufacturing, in particular to a point sol manufacturing method, device and medium. BACKGROUND

[0002] As a functional composite material, point sol fabric is widely used in medical protection, outdoor clothing and other fields by uniformly distributing glue points on the surface of the base cloth and hot pressing and bonding. In the manufacturing of point sol fabric, in order to ensure the integrity, uniformity and position accuracy of the glue points, the point sol fabric usually needs to be detected. The existing detection method is usually to detect the entire image of the point sol fabric, that is, to identify the glue points by template matching or threshold segmentation, and then to detect each glue point one by one. This method needs to traverse all pixels, and the processing speed is slow, which is difficult to meet the real-time detection demand of high-speed production line. SUMMARY

[0003] In view of the above technical problems, the technical scheme adopted by the present application is as follows: According to a first aspect of the present application, a point sol manufacturing method is provided, which comprises the following steps: S100, obtaining a target image corresponding to a point sol fabric to be detected; the target image contains a plurality of glue points; S200, dividing the target image into a plurality of sub-images according to the number of pixels corresponding to each glue point in the target image and the resolution of the target image; wherein there is a sub-image containing only one complete glue point in the plurality of sub-images; S300, performing DCT transformation on each sub-image to obtain a DCT coefficient matrix corresponding to each sub-image; S400, clustering all DCT coefficient matrices according to the characteristics of each DCT coefficient matrix to obtain a DCT coefficient matrix cluster QA containing one complete glue point; S500, processing each DCT coefficient matrix in QA to obtain a processed DCT coefficient matrix cluster WA corresponding to QA; wherein each DCT coefficient matrix in WA contains only the DCT coefficients corresponding to the image of the glue point; S600, mapping the energy center corresponding to each DCT coefficient matrix in WA to the target image to obtain the centroid of the glue point corresponding to each DCT coefficient matrix in QA; S700, splicing the sub-image corresponding to each DCT coefficient matrix not in QA in the target image to obtain the centroid of the complete glue point composed of the incomplete glue point; S800, determining whether the point sol fabric to be detected has an abnormality according to the centroid distribution of each glue point in the target image; S900, if the point sol gel fabric to be detected has no abnormality, heat pressing and bonding are performed to obtain a final fabric.

[0004] According to another aspect of the present application, a non-transitory computer readable storage medium is also provided, the storage medium storing at least one instruction or at least one program, the at least one instruction or at least one program being loaded and executed by a processor to implement the above-mentioned point sol gel fabric manufacturing method.

[0005] According to another aspect of the present application, an electronic device is also provided, comprising a processor and the above-mentioned non-transitory computer readable storage medium.

[0006] The present application has at least the following beneficial effects: The point sol gel fabric manufacturing method according to the present application divides a target image into sub-images containing complete glue points, incomplete glue points or no glue points according to the number of glue point pixels and image resolution, avoids the redundant calculation of processing a large number of pixels in the existing full image traversal detection, focuses the detection range on the glue point related area, greatly reduces the data processing amount, significantly improves the detection efficiency, and can meet the stringent requirements of real-time detection of high-speed production lines.

[0007] In addition, by DCT coefficient matrix feature clustering, the energy distribution differences of different glue point states in the frequency domain are utilized to accurately separate the coefficient matrix clusters containing complete glue points, incomplete glue points and no glue points, effectively solving the problem of complete glue point and incomplete glue point, background noise aliasing in traditional methods, and improving the purity and accuracy of glue point feature extraction; through energy center mapping and incomplete glue point splicing, only the DCT coefficients corresponding to the glue points are retained and mapped to the spatial domain to determine the centroid, and the segmented incomplete glue point sub-image is spliced and reconstructed, eliminating the interference of background noise on the centroid positioning, compensating for the positioning deviation caused by incomplete glue points or segmentation, and realizing high-precision positioning of the glue point centroid; based on multi-dimensional analysis of centroid distribution, combined with topological features and spatial pattern recognition, subtle changes in glue point completeness, distribution uniformity and position accuracy can be sensitively captured, local anomalies and small defects that are difficult to detect by traditional methods can be effectively identified, and the sensitivity and reliability of detection are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0009] Figure 1 A flowchart of the point sol gel fabric manufacturing method provided by the embodiments of the present application is shown. DETAILED DESCRIPTION

[0010] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0011] It should be noted that, based on the present disclosure, a person of ordinary skill in the art should appreciate that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, an apparatus can be implemented and / or a method can be practiced using any number of the aspects set forth herein. In addition, such an apparatus can be implemented and / or such a method can be practiced using other structure and / or functionality in addition to or other than one or more of the aspects set forth herein.

[0012] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application. Figure 1 A flowchart of a manufacturing method of a dot-sol gel fabric is shown in FIG. 1.

[0013] The manufacturing method of the dot-sol gel fabric can include the following steps. S100, obtaining a target image corresponding to a to-be-detected dot-sol gel fabric; the target image contains a plurality of glue dots.

[0014] In this embodiment, after the non-contact jet dispensing technology is used to perform the dispensing treatment on the fabric, the to-be-detected dot-sol gel fabric is obtained, and a plurality of matrix-arranged glue dots exist on the to-be-detected dot-sol gel fabric. An industrial camera can be used to capture the to-be-detected dot-sol gel fabric to obtain a target image corresponding to the to-be-detected dot-sol gel fabric. The target image can be a grayscale image. Grayscale can significantly reduce the calculation amount of subsequent DCT transformation and feature extraction.

[0015] S200, dividing the target image into a plurality of sub-images according to the number of pixels corresponding to each glue dot in the target image and the resolution of the target image; wherein there is a sub-image containing only one complete glue dot in the plurality of sub-images.

[0016] In this embodiment, the target image is segmented into a plurality of sub-images based on the number of glue dot pixels and the image resolution. Each sub-image can contain a complete glue dot, a defective glue dot, or no glue dot. Through this step, the calculation unit can be reduced, and the complexity of subsequent DCT transformation and feature analysis can be reduced. Local feature extraction is enhanced, fine analysis is performed on a single glue dot or a local area, background interference is reduced, and different size glue dots are adapted. The sub-image resolution is dynamically adjusted according to the size of the glue dot to ensure that the largest glue dot is captured completely.

[0017] Further, step S200 can include the following steps: S210, obtaining the longest side of the minimum rectangular surrounding frame corresponding to each glue point in the target image to obtain a longest side list H=(H1, H2, …, Hn). i , …, H n ), i=1, 2, …, n; wherein H i is the longest side of the rectangular surrounding frame corresponding to the i-th glue point, and n is the number of glue points.

[0018] For each glue point in the target image, the minimum rectangular surrounding frame that just wraps the glue point contour is calculated, and the longest side of the rectangular surrounding frame is extracted to obtain H.

[0019] The longest side of the minimum surrounding frame objectively reflects the geometric size of the glue point, provides a data basis for subsequent uniform setting of sub-image resolution, and avoids unreasonable sub-image division caused by size difference of glue points; the irregular shape of the glue point is converted into a rectangular feature, which is convenient for subsequent batch processing and improves the universality and robustness of the algorithm.

[0020] S220, obtaining the target longest side MH=MAX(H); MAX() is a preset maximum value function.

[0021] Taking the size of the largest glue point as a reference, the resolution of the subsequent sub-image is ensured to be sufficient to accommodate all glue points, avoiding missing of glue point information such as mutilation or truncation due to too small sub-image; in addition, taking the global maximum size as a reference facilitates subsequent uniform planning of sub-image size and reduces processing complexity caused by size difference.

[0022] S230, obtaining the pixel number NUM1 corresponding to MH.

[0023] Since MH is based on the length of image pixels, such as the number of pixels, the pixel number NUM1 corresponding to MH is directly counted, and this value is used for subsequent quantitative calculation of resolution.

[0024] Converting geometric size into image pixel unit makes the resolution setting match the actual data precision of the image and ensures the accuracy of sub-image division; NUM1 as a numerical basis provides a quantitative basis for the 2 power processing of sub-image resolution, ensuring the scientificity of size setting.

[0025] S240, if NUM1=2 r , then the resolution of the sub-image is determined to be 2 r ×2 r ; r is an integer greater than 0.

[0026] If NUM1 is exactly an integer power of 2, such as NUM1=8=2³, r=3, then the resolution of the sub-image is set to 2 r ×2r ; such as 8x8 pixels, ensuring that the resolution is square and the side length is a power of 2.

[0027] The discrete cosine transform (DCT) usually requires the input block size to be a power of 2, such as 8x8, 16x16, which can directly match the computational requirements of DCT, avoiding additional size adjustment or padding operations, and improving computational efficiency.

[0028] Ensure information integrity: a square sub-image with side length NUM1 can completely contain the largest dot, while ensuring that other smaller dots have boundary space in the sub-image, avoiding the truncation of dot edge information.

[0029] S250, if 2 r <NUM1<2 r+1 , determine the resolution of the sub-image as 2 r+1 ×2 r+1 .

[0030] If NUM1 is between two adjacent powers of 2, such as NUM1 = 10, between 2³ = 8 and , round up to the next power of 2, set the sub-image resolution to 2 r+1 ×2 r+1 , such as 16x16 pixels.

[0031] Ensure that even if the dot size is not a power of 2, the sub-image can completely contain the dot by rounding up, avoid information loss due to insufficient size, improve the robustness of the algorithm; unified computing framework, regardless of the size of the dot, the resolution of the sub-image is always a power of 2, which is convenient for subsequent batch processing of DCT transformation, such as hardware acceleration or parallel computing, and reduces the complexity of algorithm implementation; balance between computational efficiency and accuracy, avoid incomplete dot due to excessive pursuit of small size, or waste of computing resources due to excessive size, achieve optimal balance between information integrity and computational efficiency in resolution setting.

[0032] S300, DCT transform is performed on each sub-image to obtain the DCT coefficient matrix corresponding to each sub-image.

[0033] Discrete cosine transform (DCT) is performed on each sub-image to convert the spatial domain image into a frequency domain coefficient matrix. DCT transform highlights the frequency characteristics of the image, with low frequency part concentrating energy and high frequency part reflecting details.

[0034] DCT transform concentrates image energy on low frequency coefficients, suppresses noise interference, and facilitates the extraction of essential features of the dot, such as shape and texture; compared with original pixel data, DCT coefficient matrix is more compact, reducing subsequent computational complexity.

[0035] It should be noted that the skilled in the art can use the existing DCT transform method to perform DCT transform on each sub-image according to actual needs to obtain the DCT coefficient matrix corresponding to each sub-image, which is not described here.

[0036] S400, according to the characteristics of each DCT coefficient matrix, clustering all DCT coefficient matrices to obtain a DCT coefficient matrix cluster QA containing a complete glue point.

[0037] In this embodiment, feature extraction can be performed on each DCT coefficient matrix, including low-frequency energy feature extraction and non-zero coefficient distribution feature extraction.

[0038] Low-frequency energy feature extraction: After DCT transformation, the upper left corner of the matrix is the low-frequency coefficient, which represents the overall structure of the image, and the lower right corner is the high-frequency coefficient, which represents the details and edges. For sub-images containing glue points, the low-frequency energy proportion of complete glue points is higher, while the low-frequency energy proportion of incomplete glue points or no glue points is lower.

[0039] Calculate the low-frequency energy proportion of each DCT coefficient matrix: for example, take the first k x k low-frequency coefficients, such as the first 4 x 4 in an 8 x 8 matrix, calculate the energy sum and the proportion of the entire matrix energy sum; wherein the energy sum is the coefficient square sum.

[0040] Non-zero coefficient distribution feature: sub-images containing complete glue points have more concentrated and larger number of non-zero values in their DCT coefficient matrices due to their complete structure; sub-images without glue points have mostly zero or near-zero coefficients. Statistics of the number of non-zero coefficients and spatial distribution density of each matrix, such as the proportion of non-zero coefficients in high-frequency regions.

[0041] After extracting the above features, feature vector construction is performed, combining the above features such as low-frequency energy proportion, number of non-zero coefficients, high-frequency energy proportion, mean / variance of coefficient matrix, etc. into a multi-dimensional feature vector as the input of clustering.

[0042] The clustering algorithm can use K-means clustering algorithm or hierarchical clustering, i.e. merging similar clusters from bottom to top, which is convenient for clustering according to feature distance.

[0043] After clustering, the feature vector of QA satisfies the highest low-frequency energy proportion, the largest number of non-zero coefficients and the most concentrated distribution; other feature vectors clustered into QA are not the feature vectors corresponding to sub-images containing a complete glue point, such as DCT coefficient matrices containing at least one incomplete glue point and DCT coefficient matrices not containing glue points.

[0044] Further, the clustering result can be verified by manually annotating a small amount of samples, and if there is a misclassification, the feature weight or clustering center initial value can be adjusted.

[0045] S500, processing each DCT coefficient matrix in QA to obtain a processed DCT coefficient matrix cluster WA corresponding to QA; wherein each DCT coefficient matrix in WA only contains DCT coefficients corresponding to the image of the gel point.

[0046] The background noise is removed from the DCT coefficient matrix containing the complete gel point, and only the feature information of the gel point itself is retained, thereby providing pure feature data for subsequent centroid calculation.

[0047] Further, step S500 can include the following steps: S510, obtaining a standard DCT coefficient matrix D of the non-pointed gel fabric corresponding to QA bg ; wherein D bg is obtained by performing DCT transformation on the image obtained by photographing the non-pointed gel fabric; and the surface tension, fabric material and knitting method of the fabric are the same during photographing the non-pointed gel fabric and photographing the point gel fabric to be detected.

[0048] In this embodiment, the fabric without gel point, i.e., the same type of fabric without point gel processing, is photographed, and DCT transformation is performed on the image to obtain the standard DCT coefficient matrix D bg ; it should be noted that the surface tension of the fabric will change, and the change of the surface tension of the fabric will also affect the texture characteristics of the fabric. Similarly, the material and knitting method of the fabric will also affect the texture characteristics of the fabric; therefore, the surface tension, material and knitting method of the fabric during photographing are completely consistent with the fabric to be detected, so as to ensure that D bg can accurately reflect the background characteristics of the fabric to be detected, i.e., the inherent texture, structure and the like without gel point.

[0049] Through this step, a reliable background reference can be established, because D bg represents the frequency domain characteristics of the fabric without gel point, thereby providing a reference template for subsequent removal of background noise. If the photographing conditions are inconsistent, D bg may contain variables irrelevant to the background, such as fabric deformation and material difference, thereby causing errors in subsequent difference processing.

[0050] It can also improve the feature separation accuracy, by fixing the photographing conditions, to ensure that D bg is highly matched with the background characteristics of the image to be detected, so that the subsequent difference operation can accurately separate the feature signal of the gel point.

[0051] S520, for any DCT coefficient matrix D fg in QA, performing same-position difference between D fg and D bg to obtain a DCT coefficient difference matrix D fg corresponding to D bg and Ddiff =(DE1, DE2, ..., DE p , ..., DE q ), p=1,2,…,q; where DE p D fg With D bg The difference corresponding to the p-th coefficient, q is D fg With D bg The number of coefficients in the middle.

[0052] In this embodiment, the DCT coefficient matrix D for each complete glue dot in QA is... fg With standard DCT coefficient matrix D bg Subtracting the corresponding position coefficients yields the difference matrix D. diff D fg It includes background features and glue dot features, while D bg Only background features are included, therefore, D diff The essence is the glue dot characteristic, and the difference mainly reflects the contribution of the glue dots to the frequency domain coefficients.

[0053] Through the difference operation, background features, such as the frequency domain coefficients of the inherent texture of the fabric, are canceled out, and the difference matrix D... diff It focuses more on the frequency domain components corresponding to the shape, edge and other features of the glue dots, providing a foundation for subsequent noise reduction; it reduces background interference and avoids the mixing of background texture and glue dot features, so that subsequent processing can directly target the essential features of the glue dots and improve the accuracy of subsequent centroid calculation.

[0054] S530, Traverse D diff For each coefficient in the equation, if -η < DE p If <η, then D fg Set the p-th coefficient in the matrix to 0 to obtain D. fg The corresponding processed DCT coefficient matrix; η is the preset DCT coefficient difference threshold.

[0055] Set a threshold η, and traverse the difference matrix D. diff Each coefficient DE p . If DE p If the absolute value is less than η, the coefficient difference is considered to be caused by background noise or small fluctuations, rather than by glue dot characteristics; D fg The coefficients at the corresponding positions are set to 0 to eliminate the influence of background noise, and the coefficients that are finally retained correspond to the significant features of the glue dots.

[0056] η can be determined by the following method: Step 1: Collect standard data for non-sol-gel fabric: Take multiple groups of point-free sol gel fabric (consistent with the material, tension, and knitting method of the fabric to be detected), perform DCT transformation on each group of images to obtain multiple standard DCT coefficient matrices.

[0057] Step 2: Calculate the statistical distribution of background coefficients: Statistically analyze the same-position coefficients of all standard DCT coefficient matrices to calculate the mean μ and standard deviation σ of each group of same-position coefficients.

[0058] Step 3: Determine the empirical threshold of η: η can be set to k×σ, where k is an empirical coefficient and its value range is 2 to 3 to cover 95% to 99.7% of the background noise fluctuation range. For example, when k=2, η=2σ, which can filter out about 95% of the background noise coefficients.

[0059] The DCT matrix processed through the above steps only contains gel point features, ensuring that subsequent operations such as centroid calculation and anomaly detection are based on pure data, improving the quality stability and detection reliability of point sol gel fabric manufacturing.

[0060] S600, map the energy center corresponding to each DCT coefficient matrix in WA to the target image to obtain the centroid of the gel point corresponding to each DCT coefficient matrix in QA.

[0061] Further, step S600 can include the following steps: S610, for any DCT coefficient matrix D e in WA, obtain the total energy E e corresponding to D e ; where N-1 is the number of rows or columns of D e ; (u, v) is the frequency coordinate of D e ; E(u, v) is the energy of the coefficient corresponding to (u, v); E(u, v)=|D 2 (u, v)| e ; D e (u, v) is the coefficient corresponding to (u, v).

[0062] In this embodiment, for any DCT coefficient matrix D total in the sub-image set WA, calculate its total energy E total ; in signal processing, energy is positively correlated with the square of signal amplitude, and the sum of squares of DCT coefficients reflects the total energy distribution of information in the sub-image. The presence of gel points will make the coefficient energy of the corresponding region significantly higher than the background, therefore, the total energy can be used to represent the energy intensity of the gel point.

[0063] The energy concentration degree of the glue point in the frequency domain is measured by the total energy, which provides a reference for subsequent positioning; the weight of the high amplitude coefficient is amplified by the square operation, and the influence of the low energy noise is weakened, thereby improving the stability of subsequent calculation.

[0064] S620, according to E(u, v) and E total , determine the frequency coordinates (u e , v center ) of the energy center of D center ; ; ; In this embodiment, the energy center can be understood as the center of gravity of the energy distribution in the frequency domain. The coefficient energy of the glue point is higher, which will make the energy center deviate to the frequency characteristics thereof, thereby indirectly reflecting the position characteristics of the glue point in the spatial domain.

[0065] The low energy interference of the non-glue point area is eliminated by the weighted average, so that the center coordinates are closer to the actual frequency characteristics of the glue point; in addition, the energy distribution is converted into specific coordinates, which provides clear mathematical parameters for subsequent spatial domain mapping.

[0066] S630, convert (u center , v center ) into the spatial coordinates in the sub-image corresponding to D e , and then obtain the centroid of the glue point corresponding to D e .

[0067] The energy center coordinates (u center , v center ) in the frequency domain are converted into the spatial coordinates in the sub-image, thereby obtaining the centroid of the glue point.

[0068] It should be noted that the coordinate mapping relationship of the DCT transformation needs to be considered during the conversion. The DCT transformation converts the N*N image in the spatial domain into a frequency domain matrix of the same size, wherein (u, v)=(0, 0) corresponds to the direct current component, that is, the low frequency background, and the high frequency components are distributed at the edges of the matrix; the mapping of the spatial coordinates and the frequency coordinates needs to be combined with the sub-image resolution N, and the spatial centroid coordinates of the glue point in the sub-image are finally obtained through the inverse transformation logic or geometric mapping relationship, such as center-symmetric conversion.

[0069] For example: assuming that the sub-image size is N*N, the position mapping of the frequency coordinates (u, v) in the spatial domain can be achieved by deducing the basis function distribution of the inverse DCT transformation, or directly using the linear correspondence relationship between the frequency and the spatial position. Those skilled in the art can convert the energy center coordinates (u center , v center ) in the frequency domain into the spatial coordinates in the sub-image according to actual needs using the existing conversion method, which will not be described here.

[0070] By the above steps, the energy features in the frequency domain are converted into specific positions in the spatial domain, realizing the mapping from energy distribution to physical position, directly serving the glue point detection and adjustment; the calculation of the energy center combines the weights of all coefficients, which is more resistant to local noise than directly taking the extreme point, making the centroid positioning more accurate, ensuring the uniformity of the glue point distribution of the point sol gel fabric and the process precision; the accurate centroid coordinates can be used to guide the adjustment of the subsequent sol gel spraying equipment, reducing the glue point deviation error and improving the fabric quality consistency.

[0071] S700, splicing the sub-images corresponding to each DCT coefficient matrix not in the QA in the target image to obtain the centroid of the complete glue point composed of the incomplete glue points.

[0072] In this embodiment, it can be understood that since the size of the sub-image is determined based on the pixel data of the glue point, each sub-image can cover a complete glue point with a high probability, and the number of sub-images containing incomplete glue points and not containing glue points is a minority. For such sub-images, traditional image processing methods can be used to extract the centroid of the glue point.

[0073] Further, step S700 can include the following steps: S710, splicing the sub-images corresponding to each DCT coefficient matrix not in the QA in the target image to obtain the complete glue point composed of several incomplete glue points.

[0074] For each DCT coefficient matrix not in the QA, it is first converted into the corresponding sub-image, i.e., through inverse DCT transformation or direct mapping to the sub-region of the original image. Since the glue points can be distributed in multiple adjacent sub-images, for example, the glue points are captured across regions, resulting in segmentation into multiple incomplete parts, the parts belonging to the same glue point in these sub-images need to be spliced to restore the complete glue point form.

[0075] The splicing process needs to be based on the position coordinates or feature matching of the glue points, for example, using the coordinate position of the sub-image in the original image to determine the adjacent relationship, or aligning the edge features of the glue points to combine the scattered incomplete parts into a complete glue point region.

[0076] Through the above steps, it is avoided that the glue points are truncated into multiple parts due to image segmentation, ensuring that the subsequent analysis is based on the complete glue point form, reducing missed detection or misjudgment. For the glue points across the region, the splicing operation can restore their true shape, providing an accurate basis for subsequent contour extraction and centroid calculation, especially suitable for fabric detection of glue points with large size or irregular distribution.

[0077] S720, obtaining the contour of each complete glue point.

[0078] Contour extraction is performed on the complete glue dot image after splicing. The specific steps can include: Binaryzation: Separate the glue dot area from the background to form a black and white image (glue dot as foreground, background as background).

[0079] Edge detection: Use Canny operator, Sobel operator and other algorithms to extract the boundary pixels of the glue dot.

[0080] Contour tracking: Connect the edge pixels in order to form a closed contour curve, which is used to describe the shape boundary of the glue dot.

[0081] The contour is the basis for the geometric characteristics of the glue dot, such as area, perimeter, roundness, etc. Through the contour, it can be judged whether the glue dot conforms to the standard shape, such as whether there is deformation, defect, etc. At the same time, it can also distinguish adjacent glue dots. For densely distributed glue dots, contour extraction can clearly separate different glue dots, avoid false merging caused by adhesion, and improve the accuracy of detection.

[0082] S730, according to the contour of each complete glue dot, the centroid of each complete glue dot is determined.

[0083] It should be noted that those skilled in the art can use existing centroid calculation methods to determine the centroid of each complete glue dot according to actual needs, which is not described here.

[0084] The centroid can be used as a positioning reference for the glue dot on the fabric, to judge whether the glue dot distribution is uniform and whether it deviates from the preset position, to ensure the accuracy of the dot melting process; the centroid offset degree can reflect the stability of the glue dot coating, such as whether the position deviation is caused by fabric tension or process error, to provide quantitative basis for quality control in the production process; the centroid coordinates can be used for glue dot spacing calculation, arrangement rule analysis, etc., to assist in judging whether the fabric meets the design standard, such as the regularity and uniformity of the glue dot array.

[0085] S800, according to the centroid distribution of each glue dot in the target image, it is determined whether the dot melting fabric to be detected has an abnormality.

[0086] Further, step S800 can include the following steps: S810, convert the centroid of each glue dot into a two-dimensional space point set, and construct the Delaunay triangulation and Voronoi diagram corresponding to the two-dimensional space point set.

[0087] In this embodiment, each execution corresponds to a coordinate, and the centroid of each glue point can be converted into a two-dimensional point set; the spatial adjacency relationship between the glue points is captured by the Delaunay triangular network, and the Voronoi diagram quantifies the influence range of each glue point, providing a mathematical basis for subsequent topological and geometric feature extraction; the triangular network and the Voronoi diagram are based on the global topological structure, and the slight deviation of the position of a single glue point has little effect on the overall features, which is suitable for the noisy environment in the industrial scene.

[0088] S820, acquiring a topological feature corresponding to the Delaunay triangular network; the topological feature includes: the number of adjacent nodes of each centroid, the average edge length of the Delaunay triangular network, the standard deviation of the edge length, and the internal angle distribution of the triangle.

[0089] In this embodiment, the number of adjacent nodes, i.e. the number of adjacent nodes of each centroid in the triangular network, reflects the local density. The average edge length, i.e. the average value of the length of all edges of the triangular network, reflects the average distance between the glue points. The standard deviation of the edge length measures the dispersion degree of the edge length, and reflects the uniformity of the glue point distribution. The internal angle distribution of the triangle, i.e. the angle distribution of all internal angles of the triangle, evaluates the regularity of the triangular network, and in the ideal case, the internal angles of the equilateral triangle are all 60°.

[0090] The topological features are used to determine whether the arrangement of the glue points meets the design expectation, such as a uniform point array. For example, if the standard deviation is too large, there may be uneven density of glue points; abnormal internal angle distribution may indicate local disorderly arrangement. The topological features detect abnormalities from the overall structure level, avoid single-point misjudgment, and are suitable for finding batch production defects.

[0091] S830, according to the Voronoi diagram, extracting a Voronoi polygon feature corresponding to each centroid; the Voronoi polygon feature includes: the area, the perimeter and the compactness of the polygon.

[0092] In this embodiment, the polygon area is obtained by calculating the area of the Voronoi polygon corresponding to each centroid, which reflects the influence range around the glue point; the polygon perimeter, i.e. the length of the polygon boundary, indirectly reflects the adjacent complexity around the glue point; and the compactness reflects the isotropy of the glue point distribution.

[0093] The area feature directly reflects the density around the glue point, and an abnormally large or small area may correspond to a missing or overlapping glue point. The compactness feature detects the regularity of the glue point distribution, and a compactness deviating from the normal value may indicate local abnormal arrangement, such as glue point deformation caused by uneven surface tension.

[0094] S840, generating a glue point feature vector XL corresponding to the target image according to the topological feature corresponding to the Delaunay triangular network and the Voronoi polygon feature corresponding to each centroid.

[0095] All features extracted in steps S820 and S830, such as the number of adjacent nodes, average edge length, and polygon area, are combined into a high-dimensional vector XL, with each feature corresponding to one dimension of the vector.

[0096] Complex topological and geometric features are compressed into a unified vector, which facilitates subsequent similarity calculation and anomaly detection. The feature vector provides a standardized description of the glue dot distribution, which can be directly compared with a preset standard, reducing the complexity of the algorithm.

[0097] S850, if the similarity between XL and the preset standard glue dot feature vector is greater than the preset similarity threshold, then it is determined that there is no abnormality in the sol-gel fabric at the point to be detected.

[0098] The feature vector XL is compared with the preset standard adhesive dot feature vector, and the similarity is calculated, such as cosine similarity or the reciprocal of Euclidean distance. If the similarity is greater than the preset threshold T, the adhesive dot distribution of the fabric is determined to be normal; otherwise, there is an anomaly, such as missing, overlapping, or unevenly arranged adhesive dots.

[0099] By comparing feature vectors, rapid and objective anomaly detection can be achieved, avoiding the subjectivity and inefficiency of manual detection. The similarity threshold T can be dynamically adjusted according to production process requirements to adapt to different accuracy needs, such as higher thresholds required for high-end fabrics.

[0100] Furthermore, after step S830, the method may further include the following steps: S831, obtain the area of ​​the Voronoi polygon corresponding to each centroid to obtain a list of Voronoi polygon areas Y = (Y1, Y2, ..., Y...). j , ..., Y m ), j=1,2,…,m; where, Y j Let m be the area of ​​the Voronoi polygon corresponding to the j-th centroid, and m be the number of centroids.

[0101] S832, if |Y j If -λ|>θ, then the j-th centroid is determined as an abnormal centroid; where λ is the average Voronoi polygon area corresponding to Y, and θ is the preset Voronoi polygon area difference threshold.

[0102] Objective judgment rules are constructed by statistical mean and threshold to avoid the subjectivity of manual inspection; the mean-based detection method can filter the interference of individual noise points, such as accidental small errors, and focus on anomalies that significantly deviate from the overall distribution; the abnormal centroid is directly associated with the specific glue point, which facilitates subsequent traceability of the production process, such as coating head blockage, mechanical vibration, etc.

[0103] S833, if the number of abnormal centroids is greater than the preset number threshold, it is determined that the to-be-detected point sol gel fabric has an abnormality.

[0104] In this embodiment, if it is determined that the to-be-detected point sol gel fabric has an abnormality through steps S831-S833, subsequent judgment is not required, and the judgment efficiency can be further improved.

[0105] S900, if the to-be-detected point sol gel fabric has no abnormality, heat pressing and bonding are performed to obtain a final fabric.

[0106] The manufacturing method of the point sol gel fabric provided in this embodiment divides a target image into sub-images containing complete glue points, incomplete glue points or no glue points according to the number of glue point pixels and the image resolution, avoids the redundant calculation of processing a large number of pixels in the existing full-image traversal detection, focuses the detection range on the glue point related area, greatly reduces the data processing amount, significantly improves the detection efficiency, and can meet the stringent requirements of real-time detection of high-speed production lines.

[0107] In addition, through DCT coefficient matrix feature clustering, the energy distribution differences of different glue point states in the frequency domain are utilized to accurately separate the coefficient matrix clusters containing complete glue points, incomplete glue points and no glue points, effectively solving the problem of complete glue point and incomplete glue point and background noise aliasing in the traditional method, and improving the purity and accuracy of glue point feature extraction; through energy center mapping and incomplete glue point splicing, only the DCT coefficients corresponding to the glue points are retained and mapped to the spatial domain to determine the centroid, and the segmented incomplete glue point sub-image is spliced and reconstructed, eliminating the interference of background noise on the centroid positioning, compensating for the positioning deviation caused by incomplete glue points or segmentation, and realizing high-precision positioning of the glue point centroid; based on multi-dimensional analysis of centroid distribution, combined with topological features and spatial pattern recognition, subtle changes in glue point completeness, distribution uniformity and position accuracy can be sensitively captured, local abnormalities and small defects that are difficult to detect by traditional methods can be effectively identified, and the sensitivity and reliability of detection are significantly improved.

[0108] In addition, although the steps of the method in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, certain steps can be omitted, multiple steps can be combined into one step, and / or one step can be divided into multiple steps, etc.

[0109] Embodiments of the present application also provide a non-transitory computer-readable storage medium, which can be arranged in an electronic device to save at least one instruction or at least one program related to a method in the method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the method provided in the above embodiments.

[0110] The program product can employ any combination of one or more computer-readable media. The computer-readable media can be a computer-readable storage medium or a computer-readable signal medium. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0111] The computer-readable signal medium can include a computer-readable storage medium that is propagated as a carrier wave. The computer-readable signal medium can further be any computer-readable medium that is not a storage medium. The computer-readable signal medium can be a computer-readable storage medium that is a propagated signal.

[0112] The program code embodied on the computer-readable media can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0113] The program code can be executed by one or more programmable processors, which can be implemented using one or more microprocessors, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and / or any other devices suitable for retrieval and execution of instructions. The program code can execute entirely on a user's computing device, partly on the user's computing device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider (ISP).

[0114] An electronic device provided by an embodiment of the present application includes a processor and the aforementioned non-transitory computer-readable storage medium.

[0115] The electronic device is merely an example, and should not bring any limitation to the functions and usage scope of the embodiments of the present application.

[0116] The electronic device is in the form of a general purpose computing device. Components of the electronic device can include, but are not limited to, the at least one processor described above, the at least one memory described above, a bus that connects the different system components including the memory and the processor.

[0117] The memory stores a program code that can be executed by the processor, such that the processor performs the steps in the various embodiments described in this specification.

[0118] The memory can include a readable medium in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and can further include read only memory (ROM).

[0119] The memory can also include program / utility programs having a set of (at least one) program modules that include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each or a combination thereof that can include implementation of a network environment.

[0120] The bus can be representative of one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or local bus using any of a variety of bus structures.

[0121] The electronic device can also communicate with one or more external devices, such as a keyboard or a pointing device, through an I / O interface. Additionally, the electronic device can communicate with one or more devices that enable a user to interact with the electronic device, and / or one or more devices (e.g., a router, a modem, etc.) that enable the electronic device to communicate with one or more other computing devices. Such communication can occur via an I / O interface. Also, the electronic device can communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or the public network, such as the Internet) through a network adapter. The network adapter communicates with the other modules of the electronic device via the bus. Other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0122] Those skilled in the art can clearly understand the example embodiments described herein through the above description of the example embodiments, and the example embodiments described herein can be implemented by software or by software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash disk, a mobile hard disk, or the like) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to perform the methods according to the embodiments of the present disclosure.

[0123] Embodiments of the present disclosure also provide a computer program product comprising program code for causing an electronic device to perform the steps of the methods according to the various example embodiments of the present disclosure described above in the specification when the program product is run on the electronic device.

[0124] Although some specific embodiments of the present disclosure have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration, and are not intended to limit the scope of the present disclosure. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present disclosure.

Claims

1. A method of manufacturing a dot-sol gel fabric, characterized by, The method comprises the following steps: S100, acquiring a target image corresponding to a point sol gel fabric to be detected; the target image contains a plurality of gel points; S200, dividing the target image into a plurality of sub-images according to the number of pixels corresponding to each gel point in the target image and the resolution of the target image; wherein the plurality of sub-images contain a sub-image containing only one complete gel point; S300, performing DCT transformation on each sub-image to obtain a DCT coefficient matrix corresponding to each sub-image; S400, clustering all the DCT coefficient matrices according to the characteristics of each DCT coefficient matrix to obtain a DCT coefficient matrix cluster QA containing one complete gel point; S500, processing each DCT coefficient matrix in QA to obtain a processed DCT coefficient matrix cluster WA corresponding to QA; wherein each DCT coefficient matrix in WA contains only the DCT coefficients corresponding to the image of the gel point; S600, mapping the energy center corresponding to each DCT coefficient matrix in WA to the target image to obtain the centroid of the gel point corresponding to each DCT coefficient matrix in QA; S700, splicing the sub-image corresponding to each DCT coefficient matrix not in QA in the target image to obtain the centroid of the complete gel point composed of the incomplete gel point; S800, determining whether the point sol gel fabric to be detected is abnormal according to the centroid distribution of each gel point in the target image; S900, if the point sol gel fabric to be detected is not abnormal, performing hot pressing and bonding to obtain a final fabric.

2. The method of making a point-sol gel fabric according to claim 1, wherein, Step S200 comprises the following steps: S210, obtaining the longest side of the minimum rectangular surrounding frame corresponding to each glue point in the target image to obtain a longest side list H=(H1, H2, …, Hn), i=1, 2, …, n; wherein H is the longest side of the rectangular surrounding frame corresponding to the i-th glue point, and n is the number of glue points. i , …, H n n. i ​ S220, acquiring a target longest side MH=MAX(H); MAX() is a preset maximum value function; S230, acquiring the number of pixels NUM1 corresponding to MH; S240, if NUM1 = 2 r , then determine the resolution of the sub-image as 2 r × 2 r ; r is an integer greater than 0; S250, if 2 r < NUM1 < 2 r+1 , then determine the resolution of the sub-image as 2 r+1 × 2 r+1 .

3. The method of claim 1, wherein the point-sol gel fabric is formed by a process comprising: Step S500 comprises the following steps: In S510, a standard DCT coefficient matrix D of the point-free sol gel fabric corresponding to the QA is obtained bg ; wherein D bg is obtained by performing DCT transformation on an image obtained by photographing the point-free sol gel fabric; during photographing of the point-free sol gel fabric and photographing of the point sol gel fabric to be detected, the surface tension of the fabric, the fabric material, and the knitting method are the same; S520, for any DCT coefficient matrix D in QA fg , D fg and D bg , to obtain D fg , D bg and D diff = (DE1, DE2, …, DE p , …, DE q ), p = 1, 2, …, q; wherein DE p is the difference value corresponding to the pth coefficient of D fg and D bg , and q is the number of coefficients in D fg and D bg ​ S530, traversing D diff for each coefficient in D, if -η < DE p < η, set the pth coefficient in D fg to 0 to obtain D fg corresponding processed DCT coefficient matrix; and η is a preset DCT coefficient difference threshold value.

4. The method of claim 1, wherein the point-sol gel fabric is formed by a process comprising: Step S600 comprises the following steps: S610, for any DCT coefficient matrix D in the WA e , obtaining D e corresponding total energy , wherein N-1 is the number of rows or columns of D e ; (u, v) is the frequency coordinate of D e ; E(u, v) is the energy of the coefficient corresponding to (u, v); E(u, v) = |D e (u, v)| 2 ; D e (u, v) is the coefficient corresponding to (u, v) S620, based on E(u, v) and E total Determine D e Frequency coordinates of the energy center (u) center v center ); ; ; S630, convert (u center , v center ) to spatial coordinates within the corresponding sub-image, and thereby obtain the centroid of the corresponding dot. e , v center ) to spatial coordinates within the corresponding sub-image, and thereby obtain the centroid of the corresponding dot. e 5. The method of claim 1, wherein the point-sol gel fabric is formed by a process comprising: Step S700 comprises the following steps: S710, splicing the sub-image corresponding to each DCT coefficient matrix not in QA in the target image to obtain the complete gel point composed of the incomplete gel point; S720, acquiring the contour of each complete gel point; S730, determining the centroid of each complete gel point according to the contour of each complete gel point.

6. The method of claim 5, wherein the point-sol gel fabric is formed by a process comprising: Step S800 comprises the following steps: S810, converting the centroid of each gel point into a two-dimensional space point set and constructing a Delaunay triangular net and a Voronoi diagram corresponding to the two-dimensional space point set; S820, acquiring the topological features corresponding to the Delaunay triangular net; the topological features include the number of adjacent nodes of each centroid, the average edge length of the Delaunay triangular net, the standard deviation of the edge length, and the internal angle distribution of the triangle; S830, extracting the Voronoi polygon features corresponding to each centroid according to the Voronoi diagram; the Voronoi polygon features include the area, the perimeter, and the compactness of the polygon; S840, generating the gel point feature vector XL corresponding to the target image according to the topological features corresponding to the Delaunay triangular net and the Voronoi polygon features corresponding to each centroid; S850, if the similarity between the XL and the preset standard glue point feature vector is greater than the preset similarity threshold, it is determined that the point sol gel fabric to be detected does not exist abnormality.

7. The method of claim 6, wherein the point-sol gel fabric is formed by a process comprising: After step S830, the method further comprises the following steps: S831, obtain the Voronoi polygon area corresponding to each centroid to obtain a Voronoi polygon area list Y = (Y1, Y2, …, Y j ), j = 1, 2, …, m; wherein Y m is the Voronoi polygon area corresponding to the jth centroid, and m is the number of centroids. j ​ S832, if |Y j - λ > θ, the jth centroid is determined as an abnormal centroid; wherein λ is the average Voronoi polygon area corresponding to Y, and θ is a preset Voronoi polygon area difference threshold. S833, if the number of abnormal centroids is greater than the preset number threshold, it is determined that the point sol gel fabric to be detected exists abnormality.

8. A non-transitory computer-readable storage medium having stored therein at least one instruction or at least one piece of program, characterized in that, The at least one instruction or the at least one program is loaded and executed by the processor to realize the manufacturing method of the point sol gel fabric as claimed in any one of claims 1-7.

9. An electronic device, comprising: The non-transitory computer readable storage medium as claimed in claim 8 is included. The non-transitory computer readable storage medium as claimed in claim 8 is included.

Images