Ink-jet printing defect identification method and system

Abstract

The invention discloses an ink-jet printing defect identification method and system, and relates to the technical field of image processing, and the method comprises the steps: collecting a to-be-detected printing image, and carrying out the template matching with a standard image of a corresponding printing template; calculating a color difference value between each pair of matched pixel points, obtaining a color difference matrix, dividing a color difference range, and obtaining a high color difference region and a low color difference region; based on the high-color-difference area, identifying an obvious ink dot defect area; calculating the defect probability of each pixel point based on the low-color-difference region; obtaining a corrected Euclidean distance between the pixel points based on the defect probability, and clustering and identifying an unobvious ink dot defect area; the detection result and the distribution condition of the ink dot defects are counted, and feedback regulation and control of the ink jet system are achieved; through the pixel point template matching, clustering and color difference correction technologies, the problem of unobvious ink dot defects caused by missing detection is solved, and accurate identification of the ink dot defects is realized, so that the accuracy and reliability of product quality detection are enhanced, and the printing quality of products is improved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method and system for identifying defects in inkjet printing. Background Technology

[0002] Inkjet printing technology is essentially a fabric dyeing technique that utilizes the principles of a printer. It eliminates the complex plate-making process required by traditional printing, allowing direct printing onto the fabric. In an inkjet printing production scenario, digital technology controls the inkjet system, scanning and separating digital images for color separation and management. Based on the digital image data corresponding to each color at each step, different colors of ink are progressively printed onto the corresponding positions on the fabric, one color at a time, ultimately creating complex colors and patterns. Due to the compatibility between the ink and the fabric material, as well as the precision of the printhead control, ink spot defects may appear on the fabric surface. These defects directly affect the product quality grade. Therefore, it is necessary to detect ink spot defects after each step of printing the corresponding color, and to adjust the inkjet system promptly based on the detection results to ensure that the printing operation for each color meets standards, thereby ensuring the final printed fabric product is qualified.

[0003] However, when detecting ink spot defects on fabrics, existing machine vision-based quality inspection methods often employ threshold segmentation or morphological manipulation of RGB images, relying excessively on fixed or simply adaptive color difference thresholds. When dealing with complex multi-color patterns, even under constant lighting and flat fabric conditions, it is easy to miss low-contrast, inconspicuous ink spot defects. For example, light-colored ink stains on complex color textures are easily overlooked because the color difference does not reach the threshold. Moreover, missing inconspicuous ink spot defects reduces the detection accuracy, making it difficult to accurately control the inkjet system. This causes the number and area of ​​ink spot defects to gradually increase during subsequent printing of other colors, seriously affecting the final printed quality. Summary of the Invention

[0004] To improve the detection accuracy of ink dot defects and solve the problem that traditional methods easily miss subtle ink dot defects, this invention provides a method and system for identifying inkjet printing defects, the technical solution of which is as follows: In a first aspect, the present invention provides a method for identifying defects in inkjet printing, comprising the following steps: acquiring an image of the print to be detected and performing template matching with a standard image of the corresponding printing template to obtain several pairs of matching pixels; calculating the color difference value between each pair of matching pixels to obtain a color difference matrix and dividing the color difference range to obtain high color difference regions and low color difference regions; identifying obvious ink spot defect regions based on pixel data of high color difference regions; calculating the defect probability of each pixel based on pixel data of low color difference regions; obtaining the corrected Euclidean distance between pixels based on the defect probability, clustering and identifying inconspicuous ink spot defect regions based on the corrected Euclidean distance; and statistically analyzing the detection results and distribution of ink spot defects to achieve feedback control of the inkjet system. The calculation process for the defect probability is as follows: the mean of the RGB color values ​​of all pixels in the obvious ink spot defect area is obtained as the obvious defect color mean; the Euclidean distance between the RGB color value of each pixel in the low color difference area and the obvious defect color mean is taken as the ink spot color difference value; the absolute value of the difference between the color difference value of the pixel and the ink spot color difference value is taken as the corrected color difference value; the corrected color difference value is scaled by setting a hyperparameter and added to the ink spot color difference value, and then reverse mapped to obtain the correction coefficient of each pixel in the low color difference area; the product between the color difference value of the pixel and the corresponding correction coefficient is taken as the defect probability of the pixel.

[0005] Preferably, after the printing process of each color ink is completed, an image of the printed pattern to be inspected is captured; a standard image of the printing template corresponding to the current printing color in the inkjet system is obtained; the RGB color values ​​and corresponding coordinate data of all pixels in the printed pattern to be inspected and the corresponding standard image are extracted; based on the coordinate data of the pixels, the printed pattern to be inspected and the corresponding standard image are template matched to align the pixel coordinates of the two images so that the pixels of the two images correspond one-to-one, and several pairs of matching pixels are obtained.

[0006] Preferably, a nested loop structure is used to traverse all pixels on the printed image to be detected and the corresponding standard image. The distance between the RGB color values ​​of the matching pixels is calculated using the Euclidean distance formula and used as the color difference value between the matching pixels. After traversing all pixels, the color difference matrix of the printed image to be detected is obtained.

[0007] Preferably, the color difference values ​​in the color difference matrix and the RGB color values ​​and corresponding coordinate data of the pixels in the printed image to be detected are extracted. A threshold parameter is set based on the distribution of the color difference values. Pixels with color difference values ​​greater than the threshold parameter are classified as high color difference regions, and pixels with color difference values ​​less than or equal to the threshold parameter are classified as low color difference regions.

[0008] Preferably, the color difference value data, RGB color value data, and coordinate data corresponding to all pixels in the high color difference region are extracted. The coordinates, color values, and color difference values ​​corresponding to each pixel in the high color difference region are merged to construct a six-dimensional feature vector corresponding to each pixel. The six-dimensional feature vectors corresponding to all pixels in the high color difference region are standardized to obtain the normalized feature vectors corresponding to each pixel. The Euclidean distance formula is used to calculate the pairwise Euclidean distance between the normalized feature vectors of all pixels in the high color difference region. Based on the Euclidean distance, the DBSCAN clustering algorithm is executed and noise points are eliminated to obtain several effective clusters. Each effective cluster corresponds to a region with obvious ink spot defects.

[0009] Preferably, the color difference value data, RGB color value data, and coordinate data corresponding to all pixels in the low color difference region are extracted. The ink dot color difference value and the corrected color difference value corresponding to each pixel in the low color difference region are calculated. Based on the obviousness of the color difference between the ink dot color and the printing background color, a hyperparameter is set to adjust the weight of the corrected color difference value. The product of the hyperparameter and the corrected color difference value is used as the corrected color difference value. The natural exponential function is used to reverse map the sum of the ink dot color difference value and the corrected color difference value to obtain the correction coefficient of each pixel in the low color difference region. The product between the color difference value of each pixel in the low color difference region and the corresponding correction coefficient is used as the defect probability of the corresponding pixel.

[0010] Preferably, the coordinates and RGB color values ​​of each pixel within the low color difference region are merged to construct a five-dimensional feature vector corresponding to each pixel within the low color difference region. The five-dimensional feature vectors corresponding to all pixels in the low color difference region are standardized to obtain the normalized feature vector for each pixel. The original Euclidean distance between each pair of normalized feature vectors of all pixels in the low color difference region is calculated using the Euclidean distance formula. The absolute value of the difference in defect probability between each pair of pixels in the low color difference region is taken as the probability difference between the corresponding two pixels. The natural exponential function is used to positively map the probability difference between the two pixels. The product of the mapped value and the corresponding original Euclidean distance is taken as the correction value of the original Euclidean distance between the two pixels. The sum of the correction value and the corresponding original Euclidean distance is taken as the corrected Euclidean distance between the two pixels.

[0011] Preferably, based on the corrected Euclidean distance between all pixels in the low color difference region, the DBSCAN clustering algorithm is executed and noise points are eliminated to obtain several clusters. The mean value of the color difference data of all pixels in each cluster is calculated and recorded as the average color difference value. Based on the degree of color difference between the ink dot color and the printing background color, a color difference threshold T is set. When the average color difference value of a cluster is greater than T, the cluster is determined to be a region with inconspicuous ink dot defects; when the average color difference value of a cluster is less than or equal to T, the cluster is determined to be a normal region.

[0012] Preferably, based on the identification and detection results of obvious and inconspicuous ink dot defect areas, the number and area of ​​ink dot defect areas generated after the current color printing are counted, and quantity threshold parameters and area ratio threshold parameters are set to determine whether the printing quality of the current color printing is qualified; when the printing quality is determined to be unqualified, the distribution of ink dot defect areas is analyzed based on the coordinate data of all ink dot defect area pixels, and feedback control of the inkjet system is realized in the subsequent printing process.

[0013] Secondly, the present invention provides an inkjet printing defect identification system for implementing the above-mentioned inkjet printing defect identification method, comprising: a processor, a memory, a communication interface, and an image acquisition device. The processor stores computer program instructions for implementing the above-mentioned inkjet printing defect identification method. The image acquisition device is capable of adjusting and fixing the shooting angle and shooting height. The image acquisition device includes a standard light source and a high-resolution camera. The standard light source is a uniform light source with a constant color temperature. The communication interface is communicatively connected to the image acquisition device and the inkjet printing equipment.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention effectively solves the problem of missed detection of inconspicuous ink dot defects caused by the over-reliance on fixed thresholds in traditional methods by using pixel template matching, clustering, and color difference correction technologies. It can significantly improve the accuracy and reliability of ink dot defect detection in inkjet printing, which is conducive to improving the printing quality of products. Moreover, by classifying ink dot defects into obvious and inconspicuous ink dot defects, it is easy to determine the product quality level. At the same time, by analyzing the distribution of ink dot defect areas, feedback control of the inkjet system can be achieved in the subsequent printing process, thereby enabling targeted production optimization and achieving refined management of printing quality and cost. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the implementation of the inkjet printing defect identification method. Figure 2 A structural block diagram of an inkjet printing defect identification system; Detailed Implementation The technical features of the present invention will be further described in detail below with reference to the accompanying drawings so that those skilled in the art can understand them.

[0016] A method for identifying defects in inkjet printing, the implementation process is as follows: Figure 1 As shown, the specific implementation steps are as follows: Step S1: Acquire an image of the print to be detected and perform template matching with the standard image of the corresponding print template to obtain several pairs of matching pixels; Specifically, a high-resolution camera is deployed within the inkjet system, paired with a uniform light source with a constant color temperature. After the printing process of each color ink is completed, the fabric to be printed is placed flat, and the high-resolution camera captures an image of the printed fabric under the illumination of the uniform light source with a constant color temperature. Simultaneously, a standard image of the printing template corresponding to the current printed color in the inkjet system is acquired. The RGB color values ​​and corresponding coordinate data of all pixels in the printed image to be printed and the corresponding standard image are extracted. Based on the coordinate data of the pixels, the printed image to be printed and the corresponding standard image are template matched to align the pixel coordinates of the two images, so that the pixels of the two images correspond one-to-one, and several pairs of matching pixels are obtained.

[0017] In addition, the photoelectric sensor can be used to trigger the shooting synchronously with the production line. After the printing process of each color ink is completed, the same shooting angle and the same lighting and object distance are required to collect the RGB image of the printed fabric to be tested. The purpose is to eliminate the interference of lighting and position and ensure the comparability between images.

[0018] Step S2: Calculate the color difference value between each pair of matching pixels, obtain the color difference matrix, and divide the color difference range to obtain the high color difference region and the low color difference region; Specifically, a nested loop structure is used to traverse all pixels on the print image to be tested and the corresponding standard image. Since the RGB color value data of each pixel can be regarded as a three-dimensional vector, the Euclidean distance formula is used to calculate the distance between the RGB color values ​​of the matching pixels, which is used as the color difference value between the matching pixels. The color difference value is used to quantify the degree of color deviation between each pixel on the print image to be tested and the corresponding pixel on the standard image. After traversing all pixels, the color difference matrix of the print image to be tested is obtained.

[0019] In addition, the color difference values ​​in the color difference matrix and the RGB color values ​​and corresponding coordinate data of the pixels in the printed image to be detected are extracted. A threshold parameter is set based on the distribution of the color difference values. Pixels with color difference values ​​greater than the threshold parameter are divided into high color difference regions, and pixels with color difference values ​​less than or equal to the threshold parameter are divided into low color difference regions. For example, the top 70% of the pixels with color difference values ​​from largest to smallest in the color difference matrix are extracted as high color difference regions, and the remaining 30% of the pixels are divided into low color difference regions. This is used to initially screen the area range of obvious ink spot defects and the area range of potential inconspicuous ink spot defects.

[0020] Step S3: Identify areas with obvious ink spot defects based on pixel data in areas with high color difference; Specifically, the color difference value data, RGB color value data, and coordinate data corresponding to all pixels in the high color difference region are extracted. The coordinates, color values, and color difference values ​​corresponding to each pixel in the high color difference region are merged to construct a six-dimensional feature vector corresponding to each pixel. The six-dimensional feature vectors corresponding to all pixels in the high color difference region are standardized to obtain a normalized feature vector for each pixel. The Euclidean distance formula is used to calculate the pairwise Euclidean distance between the normalized feature vectors of all pixels in the high color difference region. The purpose of standardizing the six-dimensional feature vectors corresponding to all pixels is to eliminate the difference in dimensions and avoid large numerical dimensions such as coordinates from dominating the Euclidean distance, so that the features represented by each dimension in the vector contribute equally to the clustering result.

[0021] Furthermore, based on the Euclidean distance between each pair of normalized feature vectors of all pixels within the high color difference region, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm is executed, and noise points are eliminated to obtain several effective clusters. Each effective cluster corresponds to a region with a significant ink spot defect. The specific clustering process is as follows: [The following is a separate, unrelated section:] ...set according to the specific implementation scenario and the experience of the implementers... Domain radius and The value of the minimum cluster, for example: can be set The radius of the domain is 0.2. The minimum cluster size is 7, when the Euclidean distance between the feature vectors of pixels in the high color difference region is less than or equal to 7. When the corresponding pixels are considered as neighbors, several clusters are formed through density connectivity. Noise points marked as -1 during the DBSCAN clustering process are excluded, and several effective clusters can be obtained.

[0022] Step S4: Calculate the defect probability of each pixel based on the pixel data of the low color difference region; Since the visibility of the same color ink dots varies under different fabric printing background colors, color difference correction of ink dot defects on different background colors can make inconspicuous ink dot defects relatively obvious. Therefore, calculating the defect probability of each pixel in the low color difference area requires color difference correction of the pixels in the low color difference area to amplify the visibility of the ink dots. Specifically, the mean RGB color values ​​of all pixels within the obvious ink spot defect area are obtained as the obvious defect color mean. Color difference data, RGB color values, and coordinate data of all pixels within the low color difference area are extracted. The Euclidean distance formula is used to calculate the distance between the RGB color value of each pixel within the low color difference area and the obvious defect color mean, which is used as the ink spot color difference value for each pixel within the low color difference area. The smaller the color difference value of a pixel within the low color difference area, the lighter the pixel's color may be, or the closer the ink spot color is to the background color of the printing template. The absolute value of the difference between the color difference value of each pixel within the low color difference area and the ink spot color difference value is used as the corrected color difference value for each pixel. The smaller the corrected color difference value, the more likely the pixel is to be a hidden ink spot defect, and the greater the color difference value should be to highlight the difference between the ink spot color and the background color of the printing template, making inconspicuous ink spots as obvious as possible and avoiding missing inconspicuous ink spots. Furthermore, based on the degree of color difference between the ink dot color and the printing background color, a hyperparameter is set to adjust the weight of the corrected color difference value. The product of the hyperparameter and the corrected color difference value is used as the corrected color difference value. The natural exponential function is used to reverse map the sum of the ink dot color difference value and the corrected color difference value to obtain the correction coefficient of each pixel in the low color difference region. The product of the color difference value of each pixel in the low color difference region and the corresponding correction coefficient is used as the defect probability of the corresponding pixel. The defect probability represents the probability that the corresponding pixel belongs to an inconspicuous ink dot defect. The formula for calculating the defect probability of pixels in the low color difference region is as follows: In the formula, This represents the defect probability of the i-th pixel within the low color difference region. This represents the color difference value of the i-th pixel within the low color difference region. This represents the ink dot color difference value of the i-th pixel within the low color difference region. This represents the corrected color difference value of the i-th pixel within the low color difference region. This represents a hyperparameter; typically, g can be taken as an empirical value of 1.2. This represents the natural exponential function.

[0023] Step S5: Obtain the corrected Euclidean distance between pixels based on the defect probability, and cluster and identify inconspicuous ink spot defect areas based on the corrected Euclidean distance; Since maintaining the original distance relationship can protect the natural focusing of homogeneous ink dots when the defect probabilities of two pixels are similar, it is necessary to significantly increase the original distance between the two pixels when the defect probabilities of two pixels are significantly different. Even if the position coordinates of two pixels are similar, the two pixels must be forcibly separated to avoid incorrect clustering of high-probability pixels and low-probability pixels. Therefore, before clustering pixels in low color difference regions, it is necessary to correct the original Euclidean distance between pixels. Specifically, the coordinates and RGB color values ​​of each pixel within the low color difference region are merged to construct a five-dimensional feature vector for each pixel within the low color difference region. The five-dimensional feature vectors corresponding to all pixels in the low color difference region are standardized to obtain the normalized feature vector for each pixel. The original Euclidean distance between each pair of normalized feature vectors of all pixels in the low color difference region is calculated using the Euclidean distance formula. The absolute value of the difference in defect probability between each pair of pixels in the low color difference region is taken as the probability difference between the corresponding two pixels. The natural exponential function is used to positively map the probability difference between the two pixels. The product of the mapped value and the corresponding original Euclidean distance is taken as the correction value of the original Euclidean distance between the two pixels. The sum of the correction value and the corresponding original Euclidean distance is taken as the corrected Euclidean distance between the two pixels. The specific formula for calculating the corrected Euclidean distance between two pixels in the low color difference region is as follows: In the formula, This represents the corrected Euclidean distance between pixel i and pixel j within the low color difference region. This represents the original Euclidean distance between pixel i and pixel j within the low color difference region. This represents the probability difference between pixel i and pixel j within a low color difference region. This represents the defect probability of pixel i within the low color difference region. This represents the defect probability of pixel j within the low color difference region. This represents the natural exponential function.

[0024] Furthermore, based on the corrected Euclidean distance between all pixels in the low color difference region, the DBSCAN clustering algorithm is executed and noise points are eliminated to obtain several clusters. The specific clustering process is similar to that in step S3. The mean value of the color difference data of all pixels in each cluster is calculated and recorded as the average color difference value. Based on the degree of color difference between the ink dot color and the printing background color, a color difference threshold T is set, where the color difference threshold T can usually be taken as an empirical value of 20. When the average color difference value of a cluster is greater than T, the cluster is determined to be an area with inconspicuous ink dot defects. When the average color difference value of a cluster is less than or equal to T, the cluster is determined to be a normal area.

[0025] Step S6: Statistically analyze the detection results and distribution of ink droplet defects to achieve feedback control of the inkjet system; Specifically, based on the identification and detection results of obvious and inconspicuous ink spot defect areas, the number and area of ​​ink spot defect areas generated after the current color printing are counted. The ratio of the sum of the areas of all ink spot defect areas to the area of ​​the printed image to be detected is calculated to obtain the area ratio of ink spot defect areas. Quantity threshold parameters and area ratio threshold parameters are set. When the number and area ratio of ink spot defect areas do not exceed the corresponding threshold parameters, the printing quality of the current color printing is deemed qualified. When either the number or area ratio of ink spot defect areas exceeds the corresponding threshold parameters, it is deemed unqualified. When the printing quality is deemed unqualified, the distribution of ink spot defect areas is analyzed based on the coordinate data of all ink spot defect area pixels, enabling feedback control of the inkjet system in subsequent printing processes.

[0026] This invention also discloses an inkjet printing defect identification system for implementing the above-mentioned inkjet printing defect identification method, the structure of which is as follows: Figure 2 As shown, it includes: a processor, a memory, a communication interface, and an image acquisition device. The processor stores computer program instructions for implementing the above-mentioned inkjet printing defect identification method. The image acquisition device can adjust and fix the shooting angle and shooting height. The image acquisition device includes a standard light source and a high-resolution camera. The standard light source is a uniform light source with a constant color temperature. The communication interface is communicatively connected to the image acquisition device and the inkjet printing equipment.

[0027] The embodiments included in this invention are merely descriptions of preferred embodiments of the invention and are not limited to the precise structures described above and shown in the accompanying drawings. Various modifications and changes can be made without departing from the scope of protection. Any variations and improvements made by those skilled in the art to the technical solutions of this invention without departing from the design concept of this invention should fall within the scope of protection of this invention.

Claims

1. A method for identifying defects in inkjet printing, characterized in that: The image of the printed pattern to be inspected is acquired and matched with the standard image of the corresponding printing template to obtain several pairs of matching pixels. The color difference value between each pair of matching pixels is calculated to obtain a color difference matrix and the color difference range is divided to obtain high color difference area and low color difference area. Based on the pixel data of the high color difference area, obvious ink spot defect areas are identified. Calculate the defect probability of each pixel based on pixel data in low color difference regions; The corrected Euclidean distance between pixels is obtained based on the defect probability, and inconspicuous ink spot defect areas are clustered and identified based on the corrected Euclidean distance. The detection results and distribution of ink droplet defects are statistically analyzed to enable feedback control of the inkjet system. The calculation process for the defect probability is as follows: the mean of the RGB color values ​​of all pixels in the obvious ink spot defect area is obtained as the obvious defect color mean; the Euclidean distance between the RGB color value of each pixel in the low color difference area and the obvious defect color mean is taken as the ink spot color difference value; the absolute value of the difference between the color difference value of the pixel and the ink spot color difference value is taken as the corrected color difference value; the corrected color difference value is scaled by setting a hyperparameter and added to the ink spot color difference value, and then reverse mapped to obtain the correction coefficient of each pixel in the low color difference area; the product between the color difference value of the pixel and the corresponding correction coefficient is taken as the defect probability of the pixel.

2. The inkjet printing defect identification method according to claim 1, characterized in that, The process of acquiring images of the printed pattern to be inspected and matching them with standard images of the corresponding printing template includes: after the printing process of each color ink is completed, capturing images of the printed pattern to be inspected; acquiring standard images of the printing template corresponding to the current printing color in the inkjet system; extracting the RGB color values ​​and corresponding coordinate data of all pixels in the printed pattern to be inspected and the corresponding standard images; and, based on the coordinate data of the pixels, performing template matching between the printed pattern to be inspected and the corresponding standard images to align the pixel coordinates of the two images so that the pixels of the two images correspond one-to-one, thereby obtaining several pairs of matching pixels.

3. The inkjet printing defect identification method according to claim 1, characterized in that, The calculation of the color difference value between each pair of matching pixels includes: traversing all pixels on the print image to be detected and the corresponding standard image using a nested loop structure, calculating the distance between the RGB color values ​​of the matching pixels using the Euclidean distance formula, and using this distance as the color difference value between the matching pixels. After traversing all pixels, the color difference matrix of the print image to be detected is obtained.

4. The inkjet printing defect identification method according to claim 1, characterized in that, The process of dividing the color difference range includes: extracting the color difference values ​​from the color difference matrix and the RGB color values ​​and corresponding coordinate data of the pixels in the print image to be detected; setting a threshold parameter based on the distribution of the color difference values; dividing pixels with color difference values ​​greater than the threshold parameter into high color difference regions; and dividing pixels with color difference values ​​less than or equal to the threshold parameter into low color difference regions.

5. The inkjet printing defect identification method according to claim 1, characterized in that, The identification of obvious ink spot defect areas includes: extracting the color difference value data, RGB color value data, and coordinate data corresponding to all pixels in the high color difference area; merging the coordinates, RGB color values, and color difference values ​​corresponding to each pixel in the high color difference area to construct a six-dimensional feature vector corresponding to each pixel. The six-dimensional feature vectors corresponding to all pixels in the high color difference region are standardized to obtain the normalized feature vectors corresponding to each pixel. The Euclidean distance formula is used to calculate the pairwise Euclidean distance between the normalized feature vectors of all pixels in the high color difference region. Based on the Euclidean distance, the DBSCAN clustering algorithm is executed and noise points are eliminated to obtain several effective clusters. Each effective cluster corresponds to a region with obvious ink spot defects.

6. The inkjet printing defect identification method according to any one of claims 1 to 5, characterized in that, The calculation of the defect probability of each pixel includes: extracting the color difference value data, RGB color value data and coordinate data corresponding to all pixels in the low color difference region; calculating the ink dot color difference value and the corrected color difference value corresponding to each pixel in the low color difference region; setting a hyperparameter for adjusting the weight of the corrected color difference value based on the obviousness of the color difference between the ink dot color and the printing background color; using the product of the hyperparameter and the corrected color difference value as the corrected color difference value; using the natural exponential function to reverse map the sum of the ink dot color difference value and the corrected color difference value to obtain the correction coefficient of each pixel in the low color difference region; and using the product between the color difference value of each pixel in the low color difference region and the corresponding correction coefficient as the defect probability of the corresponding pixel.

7. The inkjet printing defect identification method according to claim 6, characterized in that, The step of obtaining the corrected Euclidean distance between pixels based on defect probability includes: merging the coordinates and RGB color values ​​corresponding to each pixel in the low color difference region to construct a five-dimensional feature vector corresponding to each pixel in the low color difference region. The five-dimensional feature vectors corresponding to all pixels in the low color difference region are standardized to obtain the normalized feature vector for each pixel. The original Euclidean distance between each pair of normalized feature vectors of all pixels in the low color difference region is calculated using the Euclidean distance formula. The absolute value of the difference in defect probability between each pair of pixels in the low color difference region is taken as the probability difference between the corresponding two pixels. The natural exponential function is used to positively map the probability difference between the two pixels. The product of the mapped value and the corresponding original Euclidean distance is taken as the correction value of the original Euclidean distance between the two pixels. The sum of the correction value and the corresponding original Euclidean distance is taken as the corrected Euclidean distance between the two pixels.

8. The inkjet printing defect identification method according to claim 7, characterized in that, The step of clustering and identifying inconspicuous ink spot defect areas based on modified Euclidean distance includes: performing the DBSCAN clustering algorithm and eliminating noise points based on the modified Euclidean distance between all pixels in the low color difference area to obtain several clusters; calculating the mean of the color difference values ​​of all pixels in each cluster, denoted as the average color difference value; setting a color difference threshold T based on the degree of color difference between the ink spot color and the printing background color; when the average color difference value of a cluster is greater than T, the cluster is determined to be an inconspicuous ink spot defect area; when the average color difference value of a cluster is less than or equal to T, the cluster is determined to be a normal area.

9. The inkjet printing defect identification method according to claim 1, characterized in that, The statistical analysis of ink dot defects and their distribution enables feedback control of the inkjet system. This includes: based on the identification and detection results of obvious and inconspicuous ink dot defect areas, counting the number and area of ​​ink dot defect areas generated after the current color printing, setting quantity threshold parameters and area ratio threshold parameters, and determining whether the printing quality of the current color printing is qualified; when the printing quality is determined to be unqualified, analyzing the distribution of ink dot defect areas based on the coordinate data of all ink dot defect pixels, and implementing feedback control of the inkjet system in subsequent printing processes.

10. A defect identification system for inkjet printing, characterized in that: The device includes a processor, a memory, a communication interface, and an image acquisition device. The processor stores computer program instructions for implementing the inkjet printing defect identification method according to any one of claims 1 to 9. The image acquisition device is capable of adjusting and fixing the shooting angle and shooting height. The image acquisition device includes a standard light source and a high-resolution camera. The standard light source is a uniform light source with a constant color temperature. The communication interface is communicatively connected to the image acquisition device and the inkjet printing equipment.

Images