Intelligent thermal dye sublimation transfer paper defect identification system based on image analysis

Abstract

The invention relates to the technical field of computer vision, in particular to an intelligent thermal dye sublimation transfer paper defect identification system based on image analysis. Comprising an image acquisition module, a pattern defect acquisition module, a pattern defect analysis module, a color defect acquisition module, a color defect analysis module, a paper defect detection module, a paper defect analysis module, a transfer finished product comprehensive analysis module and a management database, according to the system, a pattern evaluation coefficient of a transfer finished product image is obtained through analysis of pattern defect parameters of the transfer finished product image, a color evaluation coefficient of the transfer finished product image is obtained through analysis of color defect parameters of the transfer finished product image, and a paper evaluation coefficient of thermal dye sublimation transfer paper is obtained through analysis of paper defect parameters of the thermal dye sublimation transfer paper. Therefore, the defect evaluation index of the thermal dye sublimation transfer paper is obtained through comprehensive evaluation, specific defect sources and influence degrees in the thermal dye sublimation transfer process can be determined, and problems can be found and solved quickly.

Description

Technical Field

[0001] The invention relates to the field of computer vision technology, and in particular to an intelligent recognition system for thermal sublimation transfer paper defects based on image analysis. Background Art

[0002] Thermal sublimation transfer technology is a special printing technology. Its principle is mainly based on the fact that certain dyes can be directly transformed from solid to gas at a specific temperature and penetrate into the receiving material to achieve printing. With the continuous development of materials science and printing technology, thermal sublimation transfer technology has gradually matured and improved. Early thermal sublimation transfer was mainly used in some specific fields, such as photo printing, etc. Now it has been widely used in many industries such as textiles, plastic products, ceramics, etc. With the advancement of science and technology and changes in market demand, the requirements for transfer quality are becoming increasingly stringent.

[0003] For example, the existing Chinese patent with publication number CN116572636B discloses a detection method and system for double-roller thermal transfer equipment. This scheme can comprehensively evaluate the transfer quality of the thermal transfer equipment by detecting film defects before transfer, film transfer width, finished product defects after transfer, and waste residue after transfer, grasp the status of each link in the thermal transfer process, and promptly discover problems with the film itself and defects in the transfer process to ensure good thermal transfer effects.

[0004] However, the above patent has the following problems: the solution mainly detects the damaged positions and defect positions on the finished film material for the defects of the finished product after transfer, and does not involve the defect detection of the transfer pattern, such as blur, deformation, missing printing, etc., which may lead to an incomplete assessment of the quality of the finished product and failure to promptly discover problems with the transfer pattern, thereby affecting the overall quality and aesthetics of the product. It may also result in non-compliance with standards in some application scenarios with high requirements for pattern quality. Summary of the invention

[0005] In order to overcome the shortcomings of the background technology, the embodiment of the present invention provides a thermal sublimation transfer paper defect intelligent identification system based on image analysis, which can effectively solve the problems involved in the above background technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides an intelligent recognition system for thermal sublimation transfer paper defects based on image analysis, including: an image acquisition module for acquiring an image of thermal sublimation transfer paper, recorded as a transfer product image.

[0007] The pattern defect acquisition module is used to detect the pattern defect parameters of the transferred finished product image, and the pattern defect parameters include the degree of pattern blur, the degree of pattern deformation, and the degree of pattern missing printing.

[0008] The pattern defect analysis module is used to obtain the pattern evaluation coefficient α of the transferred finished image according to the pattern defect parameter analysis of the transferred finished image.

[0009] The color defect acquisition module is used to detect the color defect parameters of the transferred finished image. The color defect parameters include color spot ratio, color deviation degree, and color uniformity.

[0010] The color defect analysis module is used to obtain a color evaluation coefficient β of the transferred finished image according to the color defect parameter analysis of the transferred finished image.

[0011] The paper defect detection module is used to detect paper defect parameters of thermal sublimation transfer paper, and the paper defect parameters include ink accumulation degree and warping degree.

[0012] The paper defect analysis module is used to obtain the paper evaluation coefficient γ of the thermal sublimation transfer paper according to the paper defect parameter analysis of the thermal sublimation transfer paper.

[0013] The transfer product comprehensive analysis module is used to obtain the defect evaluation index of the thermal sublimation transfer paper based on the pattern evaluation coefficient, color evaluation coefficient and paper evaluation coefficient of the transfer product image, and provide feedback.

[0014] The management database is used to store pixel difference thresholds, transfer master images, and flatness difference thresholds.

[0015] Preferably, the specific analysis method of the image acquisition module is: the first step is to obtain the thermal sublimation transfer paper after transfer.

[0016] The second step is to obtain images of the thermal sublimation transfer paper after transfer through a high-definition camera.

[0017] In the third step, the obtained thermal sublimation transfer paper image is recorded as the transfer finished product image.

[0018] Preferably, the specific analysis method of the pattern defect acquisition module is as follows: the first step is to read the transferred finished image, grayscale the transferred finished image, and divide it into a number of equal-area sub-regions, which are recorded as the sub-regions of the transferred finished image, detect the grayscale value of each pixel point in each sub-region of the transferred finished image, and obtain the difference value of each pixel point in each sub-region of the transferred finished image by subtracting the grayscale value of each pixel point in each sub-region of the transferred finished image from that of its adjacent pixel point, which is recorded as Δσ im , where i represents the number of the i-th sub-region, i = 1, 2, ..., n, m represents the number of the m-th pixel, m = 1, 2, ..., q, and the preset pixel difference threshold is read from the management database, denoted as Δσ0, and substituted into the formula The pattern blur degree ξ of the transferred finished image is obtained, where n represents the number of sub-regions and q represents the number of pixels.

[0019] In the second step, the transfer master image is read from the management database, and it is adjusted to the same size as the transfer product image by scaling. Several key points are taken on the transfer master image according to the set gap, and the set key points are marked on the transfer product image in turn. The key points are matched one by one and recorded as groups of key points of the transfer image. The transfer master image is overlapped with the transfer product image, and the distance between each group of key points of the transfer image is measured, which is recorded as d j , j represents the number of the jth group of key points, j = 1, 2, ..., k, and the average distance of each group of key points in the transferred image is calculated and recorded as Substituting this into the formula The pattern deformation degree ζ of the transferred finished image is obtained, and k represents the number of key point groups.

[0020] In the third step, the transfer pattern parts in the transfer master image and the transfer product image are extracted by edge contour extraction technology, which are recorded as the transfer master pattern edge contour and the transfer product pattern edge contour. The transfer master pattern edge contour and the transfer product pattern edge contour are overlapped to obtain the number of overlapping pixels, which is recorded as N. At the same time, the total number of pixels of the transfer master image is extracted, which is recorded as N 总 , substituting it into the formula The pattern omission degree ε of the transferred finished image is obtained.

[0021] Preferably, the specific analysis method of the pattern defect analysis module is: respectively reading the pattern blur degree ξ and the pattern deformation degree of the transferred finished image. The pattern leakage degree ε is substituted into the formula The pattern evaluation coefficient α of the transferred finished image is obtained, φ1, φ2, and φ3 represent weight factors for setting the blurring degree, deformation degree, and missing degree of the pattern, respectively, and e represents a natural constant.

[0022] Preferably, the specific analysis method of the color defect acquisition module is as follows: in the first step, each pixel point in the transfer master image and the transfer finished product image is recorded as each transfer pixel point, and the chromaticity values ​​of each transfer pixel point in the transfer master image and the transfer finished product image on the three color channels of red (R), green (G), and blue (B) are detected respectively, and the difference between the two is obtained to obtain the chromaticity value difference between the transfer finished product image and each transfer pixel point in the transfer master image on the R, G, and B channels, and the difference is compared with the preset chromaticity value difference threshold value respectively, and each abnormal chromaticity pixel point is screened out, and the number of each abnormal chromaticity pixel point is counted, which is recorded as N 异常 , read the total number of pixels N of the transfer master image 总 , through the formula The color spot ratio of the transferred finished image is obtained.

[0023] The second step is to read the chromaticity value of each transferred pixel of the transfer master image. The chromaticity value of each transferred pixel of the transferred finished image The average chromaticity value of the transfer master image and the transfer finished product image is obtained by taking the average value respectively. Substituting this into the formula The color deviation degree ρ of the transferred finished image is obtained.

[0024] The third step is to read the gray value σ of each pixel in each sub-area of ​​the transferred finished image. im , through the formula Get the grayscale mean of each sub-region of the transferred finished image q represents the number of pixels, and the grayscale mean of each sub-region of the transferred finished image is averaged and recorded as Substituting this into the formula The color uniformity λ of the transferred finished image is obtained, where n represents the number of sub-regions.

[0025] Preferably, the specific analysis method of the color defect analysis module is: respectively read the color spot ratio ο, color deviation degree ρ, and color uniformity degree λ of the transferred finished image, and substitute them into the formula Get the color evaluation coefficient β of the transferred finished image, They respectively represent the weight factors of the set color spot ratio, color deviation degree, and color uniformity, and e represents a natural constant.

[0026] Preferably, the specific analysis method of the paper defect detection module is as follows: in the first step, the thermal sublimation transfer paper is stably placed on the detection platform, the detection points are selected according to the set gap, and the flatness of each detection point of the thermal sublimation transfer paper is obtained by a flatness measuring instrument, which is recorded as P f , f represents the number of the fth test point, f=1,2,...,g, the average flatness of each test point of the thermal sublimation transfer paper is calculated and recorded as And read the preset flatness difference threshold from the management database, record it as ΔP0, and substitute it into the formula The ink accumulation degree τ of the thermal sublimation transfer paper is obtained.

[0027] The second step is to place the thermal sublimation transfer paper on a flat horizontal surface, and use a vernier caliper to measure the distance between each detection point of the thermal sublimation transfer paper and the horizontal surface, which is recorded as d' f , substituting it into the formula Get the degree of warping of thermal sublimation transfer paper g represents the number of detection points.

[0028] Preferably, the specific analysis method of the paper defect analysis module is: reading the ink accumulation degree τ and the warping degree of the thermal sublimation transfer paper Substituting this into the formula The paper evaluation coefficient γ of the thermal sublimation transfer paper is obtained, η1 and η2 represent the weight factors of the set ink accumulation degree and warping degree respectively, and e represents a natural constant.

[0029] Preferably, the specific analysis method of the defect evaluation index of the thermal sublimation transfer paper is: respectively read the pattern evaluation coefficient α, the color evaluation coefficient β, and the paper evaluation coefficient γ of the transfer finished image, and substitute them into the formula Get the defect evaluation index of thermal sublimation transfer paper Wherein w1, w2, and w3 represent weight factors of the pattern evaluation coefficient, color evaluation coefficient, and paper evaluation coefficient of the thermal sublimation transfer paper, respectively.

[0030] Preferably, the specific analysis method of the comprehensive analysis module for the transfer finished product is: reading the defect evaluation index of the thermal sublimation transfer paper, comparing it with a preset defect evaluation index threshold; if the defect evaluation index of the thermal sublimation transfer paper is greater than or equal to the preset defect evaluation index threshold, it means that the defect evaluation index of the thermal sublimation transfer paper is qualified; if the defect evaluation index of the thermal sublimation transfer paper is less than the preset defect evaluation index threshold, it means that the defect evaluation index of the thermal sublimation transfer paper is unqualified, and feedback is given to the system.

[0031] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention obtains the pattern evaluation coefficient of the transferred finished image based on the pattern defect parameter analysis of the transferred finished image, which helps to timely discover and solve the pattern defect problem, thereby improving the quality stability of the transferred finished product.

[0032] 2. The present invention obtains a color evaluation coefficient of a transferred finished product image based on color defect parameter analysis of the transferred finished product image, which helps to ensure that the color of the transferred finished products produced in batches remains consistent and improves the overall quality of the product.

[0033] 3. The present invention obtains the paper evaluation coefficient of the thermal sublimation transfer paper according to the paper defect parameter analysis of the thermal sublimation transfer paper. By understanding the paper defects, targeted measures can be taken to avoid or reduce poor transfer quality caused by paper problems.

[0034] 4. The present invention obtains the defect evaluation index of the sublimation transfer paper according to the pattern evaluation coefficient, color evaluation coefficient and paper evaluation coefficient of the transferred finished product image, and provides feedback thereon, which helps to clarify the specific sources of defects and the degree of influence in the sublimation transfer process, quickly discover and solve problems, reduce unnecessary duplication of work and time waste, and improve production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 This is a system module connection diagram of the present invention.

[0037] Figure 2 for Figure 1 Schematic diagram of the method flow of the image acquisition module.

[0038] Figure 3 for Figure 1 The process judgment block diagram of the comprehensive analysis module of the transfer printing finished product. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] See also Figure 1 As shown, a thermal sublimation transfer paper defect intelligent identification system based on image analysis includes an image acquisition module, a pattern defect acquisition module, a pattern defect analysis module, a color defect acquisition module, a color defect analysis module, a paper defect detection module, a paper defect analysis module, a transfer product comprehensive analysis module, and a management database.

[0041] The management database is connected to the image acquisition module, the pattern defect analysis module, the color defect analysis module, the paper defect analysis module, and the transfer product comprehensive analysis module; the pattern defect acquisition module is connected to the pattern defect analysis module, the color defect acquisition module is connected to the color defect analysis module, the paper defect detection module is connected to the paper defect analysis module, and the transfer product comprehensive analysis module is connected to the pattern defect analysis module, the color defect analysis module, and the paper defect analysis module.

[0042] The image acquisition module is used to obtain the image of the thermal sublimation transfer paper, which is recorded as the transfer product image.

[0043] See also Figure 2 As shown, the specific analysis method of the image acquisition module is: the first step is to obtain the thermal sublimation transfer paper after transfer.

[0044] The second step is to obtain images of the thermal sublimation transfer paper after transfer through a high-definition camera.

[0045] The third step is to record the obtained thermal sublimation transfer paper image as the finished transfer product image; presenting the state after transfer in the form of an image can provide a more intuitive understanding of the situation and timely discover possible problems that may occur during the transfer process.

[0046] The pattern defect acquisition module is used to detect the pattern defect parameters of the transferred finished product image, and the pattern defect parameters include the degree of pattern blur, the degree of pattern deformation, and the degree of pattern missing printing.

[0047] The specific analysis method of the pattern defect acquisition module is as follows: the first step is to read the transferred finished image, grayscale the transferred finished image, and divide it into a number of equal-area sub-regions, which are recorded as the sub-regions of the transferred finished image, detect the grayscale value of each pixel point in each sub-region of the transferred finished image, and obtain the difference value of each pixel point in each sub-region of the transferred finished image by subtracting the grayscale value of each pixel point in each sub-region of the transferred finished image from that of its adjacent pixel point, which is recorded as Δσ im , where i represents the number of the i-th sub-region, i = 1, 2, ..., n, m represents the number of the m-th pixel, m = 1, 2, ..., q, and the preset pixel difference threshold is read from the management database, denoted as Δσ0, and substituted into the formula The pattern blur degree ξ of the transferred finished image is obtained, where n represents the number of sub-regions and q represents the number of pixels. Through grayscale processing and area division and other operations, the characteristics of the transferred finished image in different parts can be deeply understood, making the evaluation of the quality of the transferred finished product more accurate and objective, and discovering potential blur problems in time.

[0048] It should be noted that the specific analysis method for the difference values ​​of each pixel in each sub-area of ​​the transferred finished image is: read the grayscale value of each pixel in each sub-area of ​​the transferred finished image, and simultaneously obtain the grayscale values ​​of the four adjacent upper, lower, left and right pixels of each pixel in each sub-area of ​​the transferred finished image, and obtain the grayscale value difference between each pixel in each sub-area of ​​the transferred finished image and the four adjacent upper, lower, left and right pixels by subtraction, and obtain the grayscale value difference between each pixel in each sub-area of ​​the transferred finished image and the four adjacent upper, lower, left and right pixels by taking the average value, and obtain the grayscale value difference between each pixel in each sub-area of ​​the transferred finished image.

[0049] In the second step, the transfer master image is read from the management database, and it is adjusted to the same size as the transfer product image by scaling. Several key points are taken on the transfer master image according to the set gap, and the set key points are marked on the transfer product image in turn. The key points are matched one by one and recorded as groups of key points of the transfer image. The transfer master image is overlapped with the transfer product image, and the distance between each group of key points of the transfer image is measured, which is recorded as d j , j represents the number of the jth group of key points, j = 1, 2, ..., k, and the average distance of each group of key points in the transferred image is calculated and recorded as Substituting this into the formula Get the degree of pattern deformation of the transferred finished image k represents the number of key points; it can timely discover the pattern deformation problem caused by the transfer process, improve product quality, help maintain the consistency of different batches of transfer products and the master, and reduce deviations.

[0050] In the third step, the transfer pattern parts in the transfer master image and the transfer product image are extracted by edge contour extraction technology, which are recorded as the transfer master pattern edge contour and the transfer product pattern edge contour. The transfer master pattern edge contour and the transfer product pattern edge contour are overlapped to obtain the number of overlapping pixels, which is recorded as N. At the same time, the total number of pixels of the transfer master image is extracted, which is recorded as N 总 , substituting it into the formula The pattern omission degree ε of the transferred finished product image is obtained; the specific degree of pattern omission of the transferred finished product can be accurately known, which is convenient for timely discovery of problems, effectively reducing the omission phenomenon, and ensuring the integrity and quality of the transferred finished product.

[0051] The pattern defect analysis module is used to obtain the pattern evaluation coefficient α of the transferred finished image according to the pattern defect parameter analysis of the transferred finished image.

[0052] The specific analysis method of the pattern defect analysis module is: respectively reading the pattern blur degree ξ and the pattern deformation degree of the transferred finished image The pattern leakage degree ε is substituted into the formula The pattern evaluation coefficient α of the transferred finished image is obtained, φ1, φ2, and φ3 represent the weight factors of the set pattern blur, pattern deformation, and pattern missing degree, respectively, and e represents a natural constant. It can more comprehensively and accurately reflect the overall quality status of the transferred finished product, rather than considering a single dimension, making quality control more scientific and reasonable, and ensuring the stability of product quality.

[0053] It should be noted that, in a specific embodiment, φ1 can be set to 0.3, φ2 can be set to 0.3, and φ3 can be set to 0.4. Pattern omission will directly lead to incomplete pattern and seriously affect the overall effect. Therefore, the weight corresponding to pattern omission is relatively large.

[0054] The color defect acquisition module is used to detect the color defect parameters of the transferred finished image. The color defect parameters include color spot ratio, color deviation degree, and color uniformity.

[0055] The specific analysis method of the color defect acquisition module is as follows: in the first step, each pixel point in the transfer master image and the transfer finished product image is recorded as each transfer pixel point, and the chromaticity values ​​of each transfer pixel point in the transfer master image and the transfer finished product image on the three color channels of red (R), green (G), and blue (B) are detected respectively, and the difference between the two is obtained to obtain the chromaticity value difference between each transfer pixel point in the transfer finished product image and the transfer master image on the R, G, and B channels, and the difference is compared with the preset chromaticity value difference threshold, and each abnormal chromaticity pixel point is screened out, and the number of each abnormal chromaticity pixel point is counted, which is recorded as N 异常 , read the total number of pixels N of the transfer master image 总 , through the formula The color spot ratio of the transferred finished image is obtained; the color quality of the transferred finished product can be accurately understood, and color spots and other problems can be discovered in time, thereby effectively reducing defective products caused by color spots and other color problems and saving costs.

[0056] It should be noted that the specific analysis method for screening out the abnormal chromaticity pixels is: reading the chromaticity values ​​of each transferred pixel of the transfer master image and the transfer finished image on the three color channels of red (R), green (G), and blue (B), and recording the chromaticity values ​​of each transferred pixel of the transfer master image and the transfer finished image on the R channel as x represents the number of the x-th transferred pixel, x=1,2,...,y, through the formula Get the R channel chromaticity value difference of each transferred pixel point between the transferred finished image and the transferred master image Denoted as the R channel chromaticity value difference of each transferred pixel point, according to the method of analyzing the R channel chromaticity value difference of each transferred pixel point, the G channel chromaticity value difference and the B channel chromaticity value difference of each transferred pixel point are analyzed in turn, and they are compared with the set chromaticity value difference thresholds respectively. If the chromaticity value difference of a certain transferred pixel point on any channel is greater than the set chromaticity value difference threshold, the transferred pixel point is determined to be an abnormal chromaticity pixel point. If the chromaticity value differences of a certain transferred pixel point on the R, G, and B channels are all less than or equal to the set chromaticity value difference threshold, the chromaticity value difference of the transferred pixel point is determined to be qualified.

[0057] It should be noted that, in a specific embodiment, the chromaticity value difference thresholds set for the three color channels of red (R), green (G), and blue (B) are 5, 3, and 4, respectively. For a certain transferred pixel, its chromaticity value difference in the R channel is 6, its chromaticity value difference in the G channel is 2, and its chromaticity value difference in the B channel is 5. Since the chromaticity value differences of the transferred pixel in the R channel and the B channel exceed their respective corresponding chromaticity value difference thresholds, the transferred pixel is determined to be an abnormal chromaticity pixel.

[0058] The second step is to read the chromaticity value of each transferred pixel of the transfer master image. The chromaticity value of each transferred pixel of the transferred finished image The average chromaticity value of the transfer master image and the transfer finished product image is obtained by taking the average value respectively. Substituting this into the formula The color deviation degree ρ of the transferred finished image is obtained; the color difference can be reduced, the color consistency between the transferred finished image and the master is ensured to be high, and the color stability and consistency of the transferred product is maintained.

[0059] The third step is to read the gray value σ of each pixel in each sub-area of ​​the transferred finished image. im , through the formula Get the grayscale mean of each sub-region of the transferred finished image q represents the number of pixels, and the grayscale mean of each sub-region of the transferred finished image is averaged and recorded as Substituting this into the formula The color uniformity λ of the transferred finished image is obtained, where n represents the number of sub-regions; the uniformity of the color distribution of the transferred finished image in each region can be accurately understood, which helps to timely discover possible local color unevenness problems.

[0060] The color defect analysis module is used to obtain a color evaluation coefficient β of the transferred finished image according to the color defect parameter analysis of the transferred finished image.

[0061] The specific analysis method of the color defect analysis module is: respectively read the color spot ratio ο, color deviation degree ρ, and color uniformity degree λ of the transferred finished image, and substitute them into the formula Get the color evaluation coefficient β of the transferred finished image, They represent the weighting factors of the set color spot ratio, color deviation degree, and color uniformity respectively, and e represents a natural constant. By comprehensively considering the performance of the transferred finished image in multiple key color aspects, it helps to maintain the color stability and consistency of the product.

[0062] It should be noted that, in a specific embodiment, It can be set to 0.4. It can be set to 0.4. It can be set to 0.2. The proportion of color spots is relatively critical. If the proportion of color spots is too large, it will have a very significant adverse effect on the overall color performance. The degree of color deviation is also relatively important. It will make the color appear inconsistent with expectations and have a greater impact on the visual effect. Although color uniformity is also important, it may have a smaller impact in comparison. However, if the color uniformity is very poor, it will also seriously reduce the quality of the transferred product. Therefore, the weights corresponding to the proportion of color spots and the degree of color deviation are relatively high.

[0063] The paper defect detection module is used to detect paper defect parameters of thermal sublimation transfer paper, and the paper defect parameters include ink accumulation degree and warping degree.

[0064] The specific analysis method of the paper defect detection module is as follows: In the first step, the thermal sublimation transfer paper is placed stably on the detection platform, and the detection points are selected according to the set gap. The flatness of each detection point of the thermal sublimation transfer paper is obtained by a flatness measuring instrument, which is recorded as P f , f represents the number of the fth test point, f=1,2,...,g, the average flatness of each test point of the thermal sublimation transfer paper is calculated and recorded as And read the preset flatness difference threshold from the management database, record it as ΔP0, and substitute it into the formula The ink accumulation degree τ of the thermal sublimation transfer paper is obtained; it can ensure the flatness of the paper, help to achieve better transfer effects, and avoid problems such as uneven ink distribution caused by unevenness. When the flatness of the paper is good, the ink will be distributed relatively more evenly on the surface of the paper, and it is not easy to have excessive or insufficient local accumulation, which helps to achieve good printing or transfer effects. If the paper is uneven and has raised or depressed areas, during the printing or transfer process, the ink may accumulate more on the raised areas and less on the depressed areas, resulting in uneven ink distribution, affecting the quality and clarity of the pattern or text, and may even cause problems such as different shades of ink and ghosting. At the same time, uneven paper may cause uneven pressure distribution during printing or transfer, further affecting the uniformity and stability of ink accumulation.

[0065] The second step is to place the thermal sublimation transfer paper on a flat horizontal surface, and use a vernier caliper to measure the distance between each detection point of the thermal sublimation transfer paper and the horizontal surface, which is recorded as d' f , substituting it into the formula Get the degree of warping of thermal sublimation transfer paper g indicates the number of detection points; it can promptly detect whether the transfer paper has warping problems, ensure product quality, help understand the performance of the transfer paper under different conditions, and reduce errors in subsequent processing or use caused by warping.

[0066] The paper defect analysis module is used to obtain the paper evaluation coefficient γ of the thermal sublimation transfer paper according to the paper defect parameter analysis of the thermal sublimation transfer paper.

[0067] The specific analysis method of the paper defect analysis module is: reading the ink accumulation degree τ and the warping degree of the thermal sublimation transfer paper Substituting this into the formula The paper evaluation coefficient γ of thermal sublimation transfer paper is obtained, where η1 and η2 represent the weight factors of the set ink accumulation degree and warping degree respectively, and e represents a natural constant. It can accurately reflect the comprehensive condition of the paper in terms of ink accumulation and warping, rather than considering a single indicator, which is helpful to strictly control the paper quality during the production process and discover problems in time.

[0068] It should be noted that in a specific embodiment, η1 can be set to 0.6 and η2 can be set to 0.4. Ink accumulation will directly affect the effect and quality of transfer, and may cause problems such as unclear patterns and uneven colors. The impact of the degree of warping may be relatively small, but if the warping is severe, it will also affect the use and transfer operations to a certain extent. Therefore, the weight corresponding to the degree of ink accumulation is higher.

[0069] The comprehensive analysis module for the finished transfer product is used to analyze and obtain the defect evaluation index of the thermal sublimation transfer paper according to the pattern evaluation coefficient, color evaluation coefficient and paper evaluation coefficient of the finished transfer product image, and to provide feedback thereon.

[0070] See also Figure 3 As shown, the specific analysis method of the defect evaluation index of the thermal sublimation transfer paper is: read the pattern evaluation coefficient α, color evaluation coefficient β, and paper evaluation coefficient γ of the transferred finished image respectively, and substitute them into the formula Get the defect evaluation index of thermal sublimation transfer paper Among them, w1, w2, and w3 represent the weight factors of the pattern evaluation coefficient, color evaluation coefficient, and paper evaluation coefficient of the set transfer product image, respectively; the defect evaluation index can more accurately discover and locate possible quality defects, ensure the stable quality of the produced transfer products, and produce higher quality transfer products.

[0071] It should be noted that, in a specific embodiment, w1 can be set to 0.5, w2 can be set to 0.3, and w3 can be set to 0.2. The pattern evaluation coefficient may be more important because the integrity and clarity of the pattern directly affect the appearance and quality of the finished product. The color evaluation coefficient is also very critical. Accurate color is crucial to the quality of the finished product. The paper evaluation coefficient of thermal sublimation transfer paper is relatively less important, but it is still not to be ignored. Therefore, the weight corresponding to the pattern evaluation coefficient of the transferred finished product image is higher, the weight corresponding to the color evaluation coefficient is second, and the weight corresponding to the paper evaluation coefficient of thermal sublimation transfer paper is lower.

[0072] The specific analysis method of the comprehensive analysis module of the transfer finished product is: reading the defect evaluation index of the thermal sublimation transfer paper, comparing it with the preset defect evaluation index threshold, if the defect evaluation index of the thermal sublimation transfer paper is greater than or equal to the preset defect evaluation index threshold, it means that the defect evaluation index of the thermal sublimation transfer paper is qualified, if the defect evaluation index of the thermal sublimation transfer paper is less than the preset defect evaluation index threshold, it means that the defect evaluation index of the thermal sublimation transfer paper is unqualified, and feedback is given to the system; qualified and unqualified products can be effectively screened out to ensure that product quality meets certain standards, and production links can be optimized in a targeted manner according to feedback information to improve overall production efficiency and quality.

[0073] The management database is used to store pixel difference thresholds, transfer master images, and flatness difference thresholds.

[0074] The present invention obtains a pattern evaluation coefficient of a transferred finished image according to pattern defect parameter analysis of the transferred finished image, obtains a color evaluation coefficient of the transferred finished image according to color defect parameter analysis of the transferred finished image, obtains a paper evaluation coefficient of the thermal sublimation transfer paper according to paper defect parameter analysis of the thermal sublimation transfer paper, and then obtains a defect evaluation index of the thermal sublimation transfer paper through comprehensive evaluation, and provides feedback thereon, which helps to clarify the specific defect sources and influence degrees in the thermal sublimation transfer process, quickly discover and solve problems, reduce unnecessary duplication of work and time waste, and improve production efficiency.

[0075] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.

Claims

1. An intelligent identification system for thermal sublimation transfer paper defects based on image analysis, characterized in that: The system specifically includes the following modules: An image acquisition module is used to obtain an image of the thermal sublimation transfer paper, which is recorded as a finished transfer product image; The pattern defect acquisition module is used to detect the pattern defect parameters of the transferred finished product image, and the pattern defect parameters include the degree of pattern blur, the degree of pattern deformation, and the degree of pattern missing printing; A pattern defect analysis module, used for analyzing the pattern defect parameters of the transferred finished image to obtain a pattern evaluation coefficient α of the transferred finished image; The color defect acquisition module is used to detect the color defect parameters of the transferred finished product image. The color defect parameters include the color spot ratio, color deviation degree, and color uniformity; A color defect analysis module is used to obtain a color evaluation coefficient β of the transferred finished image according to the color defect parameter analysis of the transferred finished image; The paper defect detection module is used to detect the paper defect parameters of the thermal sublimation transfer paper, including the degree of ink accumulation and the degree of warping; A paper defect analysis module is used to obtain a paper evaluation coefficient γ of the thermal sublimation transfer paper according to the paper defect parameter analysis of the thermal sublimation transfer paper; The transfer product comprehensive analysis module is used to analyze and obtain the defect evaluation index of the thermal sublimation transfer paper according to the pattern evaluation coefficient, color evaluation coefficient and paper evaluation coefficient of the transfer product image, and provide feedback thereon; The management database is used to store pixel difference thresholds, transfer master images, and flatness difference thresholds.

2. According to claim 1, the thermal sublimation transfer paper defect intelligent identification system based on image analysis is characterized in that: The specific analysis method of the image acquisition module is: The first step is to obtain the thermal sublimation transfer paper after transfer; The second step is to obtain images of the thermal sublimation transfer paper after transfer through a high-definition camera; In the third step, the obtained thermal sublimation transfer paper image is recorded as the transfer finished product image.

3. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 2, characterized in that: The specific analysis method of the pattern defect acquisition module is: The first step is to read the finished transfer image, grayscale the finished transfer image, and divide it into several equal-area sub-regions, which are recorded as the sub-regions of the finished transfer image. The grayscale value of each pixel in each sub-region of the finished transfer image is detected, and the difference between each pixel in each sub-region of the finished transfer image and its adjacent pixel is obtained by subtracting the grayscale value of each pixel in each sub-region of the finished transfer image, which is recorded as Δσ im , where i represents the number of the i-th sub-region, i = 1, 2, ..., n, m represents the number of the m-th pixel, m = 1, 2, ..., q, and the preset pixel difference threshold is read from the management database, denoted as Δσ0, and substituted into the formula Obtain the pattern blur degree ξ of the transferred finished image, where n represents the number of sub-regions and q represents the number of pixels; In the second step, the transfer master image is read from the management database, and it is adjusted to the same size as the transfer product image by scaling. Several key points are taken on the transfer master image according to the set gap, and the set key points are marked on the transfer product image in turn. The key points are matched one by one and recorded as groups of key points of the transfer image. The transfer master image is overlapped with the transfer product image, and the distance between each group of key points of the transfer image is measured, which is recorded as d j , j represents the number of the jth group of key points, j = 1, 2, ..., k, and the average distance of each group of key points in the transferred image is calculated and recorded as Substituting this into the formula Get the degree of pattern deformation of the transferred finished image k represents the number of key point groups; In the third step, the transfer pattern parts in the transfer master image and the transfer product image are extracted by edge contour extraction technology, which are recorded as the transfer master pattern edge contour and the transfer product pattern edge contour. The transfer master pattern edge contour and the transfer product pattern edge contour are overlapped to obtain the number of overlapping pixels, which is recorded as N. At the same time, the total number of pixels of the transfer master image is extracted, which is recorded as N 总 , substituting it into the formula The pattern omission degree ε of the transferred finished image is obtained.

4. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 3 is characterized in that: The specific analysis method of the pattern defect analysis module is: Read the blur and deformation of the transferred image. The pattern leakage degree ε is substituted into the formula The pattern evaluation coefficient α of the transferred finished image is obtained, φ1, φ2, and φ3 represent weight factors for setting the blurring degree, deformation degree, and missing degree of the pattern, respectively, and e represents a natural constant.

5. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 1, characterized in that: The specific analysis method of the color defect acquisition module is: In the first step, each pixel in the transfer master image and the transfer finished image is recorded as each transfer pixel, and the chromaticity values ​​of each transfer pixel in the transfer master image and the transfer finished image on the three color channels of red (R), green (G), and blue (B) are detected respectively, and the difference between the two is obtained to obtain the chromaticity value difference between each transfer pixel in the transfer finished image and the transfer master image on the R, G, and B channels, and the difference is compared with the preset chromaticity value difference threshold, and each abnormal chromaticity pixel is screened out, and the number of each abnormal chromaticity pixel is counted, which is recorded as N 异常 , read the total number of pixels N of the transfer master image 总 , through the formula Obtaining the color spot ratio of the transferred finished image; The second step is to read the chromaticity value of each transferred pixel of the transfer master image. The chromaticity value of each transferred pixel of the transferred finished image The average chromaticity value of the transfer master image and the transfer finished product image is obtained by taking the average value respectively. Substituting this into the formula Obtain the color deviation degree ρ of the transferred finished image; The third step is to read the gray value σ of each pixel in each sub-area of ​​the transferred finished image. im , through the formula Get the grayscale mean of each sub-region of the transferred finished image q represents the number of pixels, and the grayscale mean of each sub-region of the transferred finished image is averaged and recorded as Substituting this into the formula The color uniformity λ of the transferred finished image is obtained, where n represents the number of sub-regions.

6. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 5, characterized in that: The specific analysis method of the color defect analysis module is: Read the color spot ratio ο, color deviation degree ρ, and color uniformity λ of the transferred finished image respectively, and substitute them into the formula Get the color evaluation coefficient β of the transferred finished image, They respectively represent the weight factors of the set color spot ratio, color deviation degree, and color uniformity, and e represents a natural constant.

7. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 1, characterized in that: The specific analysis method of the paper defect detection module is: The first step is to place the thermal sublimation transfer paper steadily on the testing platform, select the testing points according to the set gap, and obtain the flatness of each testing point of the thermal sublimation transfer paper through the flatness measuring instrument, which is recorded as P f , f represents the number of the fth test point, f=1,2,...,g, the average flatness of each test point of the thermal sublimation transfer paper is calculated and recorded as And read the preset flatness difference threshold from the management database, record it as ΔP0, and substitute it into the formula Obtain the ink accumulation degree τ of the thermal sublimation transfer paper; The second step is to place the thermal sublimation transfer paper on a flat horizontal surface, and use a vernier caliper to measure the distance between each detection point of the thermal sublimation transfer paper and the horizontal surface, which is recorded as d' f , substituting it into the formula The warping degree l of the thermal sublimation transfer paper is obtained, and g represents the number of detection points.

8. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 7, characterized in that: The specific analysis method of the paper defect analysis module is as follows: Read the ink accumulation degree τ and warping degree l of the thermal sublimation transfer paper and substitute them into the formula The paper evaluation coefficient γ of the thermal sublimation transfer paper is obtained, η1 and η2 represent the weight factors of the set ink accumulation degree and warping degree respectively, and e represents a natural constant.

9. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 1, characterized in that: The specific analysis method of the transfer finished product comprehensive analysis module is: Read the pattern evaluation coefficient α, color evaluation coefficient β, and paper evaluation coefficient γ of the transfer finished image respectively, and substitute them into the formula Get the defect evaluation index of thermal sublimation transfer paper Wherein w1, w2, and w3 represent weight factors of the pattern evaluation coefficient, color evaluation coefficient, and paper evaluation coefficient of the thermal sublimation transfer paper, respectively.

10. The thermal sublimation transfer paper defect intelligent identification system based on image analysis according to claim 9, characterized in that: The specific analysis method of the transfer finished product comprehensive analysis module is: The defect evaluation index of the thermal sublimation transfer paper is read and compared with the preset defect evaluation index threshold. If the defect evaluation index of the thermal sublimation transfer paper is greater than or equal to the preset defect evaluation index threshold, it means that the defect evaluation index of the thermal sublimation transfer paper is qualified. If the defect evaluation index of the thermal sublimation transfer paper is less than the preset defect evaluation index threshold, it means that the defect evaluation index of the thermal sublimation transfer paper is unqualified, and feedback is given to the system.

Images