Code spraying quality detection method of intelligent code spraying machine

Abstract

The invention relates to the technical field of image processing, in particular to a code spraying quality detection method of an intelligent code spraying machine. The method comprises the following steps: collecting a code spraying area image, obtaining a binary image through dual histogram equalization and binarization processing, carrying out limit connected domain extraction on the binary image, carrying out simulation reconstruction to generate a code spraying model, carrying out skeleton splitting on the simulation model to obtain a basic skeleton, analyzing the extension coherence of the skeleton, and obtaining a code spraying result. The method comprises the steps of generating stripe continuity quality, performing gap dispersion and uniformity evaluation on a simulation model based on a code spraying skeleton to obtain stripe uniformity quality, deducing readability of a sprayed code according to the stripe continuity and uniformity quality, performing quality grade mapping, and generating a comprehensive quality evaluation report. The code spraying quality detection method is more comprehensive and more reliable.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for detecting the inkjet quality of an intelligent inkjet printer. Background Art

[0002] In modern industrial production, the application scope of inkjet coding technology is wide, covering the real-time printing of production batches, dates and identification information. This identification information is often directly attached to the surface of products or packaging through inkjet printers for production tracking and anti-counterfeiting. However, due to the fast production speed and complex environmental conditions, the quality of inkjet coding fluctuates, and the content of the inkjet coding may be blurred, detached or incomplete. In order to ensure the clarity and durability of product information, inkjet coding quality detection technology has received more and more attention. The traditional manual detection method can no longer meet the needs of efficient and accurate quality control. The current inkjet coding quality detection methods mostly rely on image processing technology, such as binarization and connected domain analysis of inkjet coding images to identify the integrity of characters and the continuity of stripes. However, the accuracy of inkjet coding detection in the existing technology is limited by the image preprocessing effect. The image quality under different lighting conditions and complex backgrounds will affect the extraction and analysis accuracy of the inkjet coding area. In addition, a single image processing method is difficult to effectively realize the multi-angle evaluation of details such as the uniformity, coherence and gap distribution of the inkjet coding stripes, and it is impossible to comprehensively and accurately judge the quality of the inkjet coding. Summary of the invention

[0003] Based on this, it is necessary to provide a method for detecting the inkjet quality of an intelligent inkjet printer to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for detecting the inkjet quality of an intelligent inkjet printer comprises the following steps: Step S1: collecting an image of the coding area; performing double histogram equalization on the image of the coding area, and performing binary segmentation to obtain a binary image; Step S2: extracting the extreme connected domain of the binary image, and reconstructing the coding simulation to generate a simulation coding model; Step S3: splitting the stripe skeleton of the simulated inkjet coding model to obtain the basic skeleton of the inkjet coding; performing extension and coherence analysis on the basic skeleton of the inkjet coding to generate stripe continuity quality; Step S4: Based on the basic skeleton of inkjet printing, gap dispersion analysis is performed on the simulated inkjet printing model, and uniformity evaluation is performed to obtain stripe uniformity quality; Step S5: Deducing the readable quality of the simulated inkjet coding model according to the stripe continuity quality and the stripe uniformity quality, and performing quality grade mapping to generate a comprehensive quality assessment report.

[0005] The present invention adjusts the illumination and contrast of the image through double histogram equalization to ensure that the image detail information is clearly captured under different illumination conditions, and improves the contrast of the coding area in the image through binary segmentation, further highlights the edge contour of the character, and facilitates subsequent feature extraction. The binary image is subjected to extreme connected domain extraction to effectively remove noise interference, ensure that the character structure in the coding area is complete and clear, and restores the coding form through coding simulation reconstruction to generate a simulated coding model with a clear structure without affecting the content of the coding characters, providing image data with a complete structure for subsequent quality inspection, performing stripe skeleton splitting on the simulated coding model, retaining the main contour of the character, removing redundant information, and obtaining the basic coding skeleton. The analysis can accurately reflect the continuity of characters and generate the quality of stripe continuity, so that the continuity information between the coded characters can be clearly displayed, providing reliable data for evaluating the clarity and recognizability of the coding. The gap discreteness analysis of the simulated coding model is performed based on the basic skeleton of the coding. By evaluating the distribution of gaps, it is ensured that there are no obvious breaks and uniformity on the edges of the characters, and the presence of gaps and defects in the characters is accurately identified. The stripe uniformity quality is generated, and the readability quality of the simulated coding model is deduced according to the stripe continuity quality and stripe uniformity quality. The quality assessment results are visualized through quality grade mapping, and a comprehensive quality assessment report is generated to ensure the readability of the coded characters under different lighting, clarity and uniformity conditions, and provide quantifiable coding quality assessment results.

[0006] Preferably, step S1 comprises the following steps: Step S11: collecting the image of the coding area; performing global histogram equalization processing on the image of the coding area to obtain an enhanced coding image; Step S12: performing non-enhanced area detection on the enhanced inkjet coding image to obtain a non-uniform area; Step S13: performing local adaptive equalization processing on the non-uniform area to generate a double equalized image; Step S14: performing maximum inter-class variance analysis on the double equalized image to obtain a cutting variance threshold parameter; Step S15: performing binarization segmentation on the double equalized image based on the cutting variance threshold parameter to obtain a binarized image.

[0007] The present invention improves the contrast of the overall image through global enhancement, highlights the structural details of the coding area, makes the brightness and contrast of the image evenly distributed in the entire coding area, obtains an enhanced coding image, improves the recognizability of the coding, and facilitates subsequent processing. The enhanced coding image is detected in the non-enhanced area, and the areas in the image that are not uniformly enhanced are accurately identified. By locating these areas, smooth processing of details is ensured to obtain uneven areas, provide accurate regional information for subsequent local enhancement, improve the accuracy of detection, and perform local adaptive equalization processing on the uneven areas to solve the problem of local details missed by global equalization. , optimize the brightness and contrast of different areas through adaptive equalization to achieve double equalization effect, perform maximum inter-class variance analysis on the double equalized image, and calculate the optimal cutting variance threshold parameter to ensure clear separation of the background and foreground of the coding area, provide accurate image segmentation basis, and effectively improve the accuracy of binary segmentation. Binarize the double equalized image based on the cutting variance threshold parameter, and obtain a clear binary image through high-precision binarization processing, ensuring that the edges of the characters in the coding area are clear and without breaks, and the segmentation effect is stable and consistent, providing accurate basic data for subsequent coding quality inspection.

[0008] Preferably, step S2 comprises the following steps: Step S21: Perform eight-neighborhood channel statistics on the binary image to obtain a candidate connected domain; Step S22: Tracing the boundary contour of the candidate connected domain to generate a connected domain contour point set; Step S23: identifying extreme points of the connected domain contour point set to obtain key contour points of the connected domain; Step S24: performing feature topological relationship analysis on key contour points of the connected domain to generate a contour point topological structure; Step S25: reconstructing the key contour points of the connected domain based on the contour point topological structure to obtain the key contour for inkjet printing; Step S26: Fill the key outline of the inkjet coding with inkjet texture according to the binary image to generate a simulated inkjet coding model.

[0009] The present invention performs eight-neighborhood channel statistics on the binary image, accurately analyzes the pixel connectivity in the image, effectively identifies all candidate connected areas in the image, ensures the coherence and integrity of the coding area, forms a candidate connected domain, tracks the boundary contour of the candidate connected domain, obtains the accurate coordinates of the boundary through continuous tracking of contour points, generates a connected domain contour point set, ensures accurate boundary delineation of the coding area, obtains complete boundary information, identifies extreme points on the connected domain contour point set, determines key boundary features by extracting the geometric extreme points of the contour, obtains key contour points of the connected domain, accurately captures key nodes of the coding boundary, ensures the effectiveness and representativeness of the contour points, and features the key contour points of the connected domain. Topological relationship analysis accurately establishes the spatial structural relationship between each contour point, generates the contour point topological structure, ensures that the relative position of the coding boundary features in space remains consistent, reconstructs the contours of the key contour points in the connected domain based on the contour point topological structure, and guides the overall reconstruction of the contour through the topological structure to obtain the key contour of the coding, ensures the continuity and accuracy of the reconstructed contour, improves the boundary expression accuracy of the coding model, fills the key contour of the coding with the coding texture according to the binary image, restores the real texture of the coding through the filling operation, generates a simulated coding model, ensures the texture consistency and visual realism of the simulation model, provides a simulated coding image that is closer to the actual effect, and provides accurate basic data for coding quality detection.

[0010] Preferably, step S3 comprises the following steps: Step S31: positioning the contour branch points of the simulation inkjet model to obtain inkjet branch position information; Step S32: splitting the simulated inkjet model into skeletons based on the inkjet branch position information to generate an inkjet basic skeleton; Step S33: monitoring the extension breakpoints of the coding basic skeleton to obtain the skeleton breakpoint positions; Step S34: performing a coherence analysis on the basic skeleton of the coding based on the skeleton breakpoint positions to generate a skeleton continuity index; Step S35: mapping the skeleton continuity index to a quality level according to a preset continuity quality reference value to generate a stripe continuity quality.

[0011] The present invention obtains the branch position information of the coding model by accurately identifying the branch nodes in the coding model, ensures the integrity of the complex contour structure of the coding model, improves the accuracy of the model branch information, performs skeleton splitting on the simulated coding model based on the coding branch position information, generates a basic coding skeleton, guides the skeleton splitting process through branch nodes, ensures the continuity and clarity of the skeleton, helps to accurately display the overall shape of the coding area, improves the structural clarity of the coding model, performs extended breakpoint monitoring on the basic coding skeleton, accurately locates the breakpoint existence area by detecting the breakpoint position in the skeleton, obtains the skeleton breakpoint position, ensures the integrity and continuity recognition of the skeleton structure, and improves the breakpoint The positioning accuracy of the point data is analyzed based on the continuity analysis of the basic skeleton of the coding based on the skeleton breakpoint position. The continuity of the skeleton in the coding area is obtained by analyzing the skeleton breakpoint position, and the skeleton continuity index is generated to ensure the quantitative analysis of the coding quality in terms of structural continuity, improve the evaluation ability of the structural consistency of the coding model, and map the quality level of the skeleton continuity index according to the preset continuity quality reference value to generate the stripe continuity quality. The skeleton continuity quality level is mapped by comparing with the reference value to ensure that the coding skeleton meets the quality requirements in terms of continuity, effectively evaluate the structural integrity and uniformity of the coding, and provide stripe structure quality data for the final coding quality evaluation.

[0012] Preferably, step S34 includes the following steps: The basic coding skeleton is segmented based on the skeleton breakpoint positions to obtain a multi-segment coding skeleton; Compare the adjacent length differences of multiple sections of inkjet skeletons to generate the adjacent inkjet length differences; Compare the adjacent angle differences of multiple sections of inkjet skeletons to obtain the adjacent inkjet angle differences; Perform segment continuity evaluation on the difference in adjacent code lengths and adjacent code angles to generate segment continuity quality; Based on the segmented continuity quality, the coding basic skeleton is mapped to the continuity index to generate the skeleton continuity index.

[0013] The present invention realizes the detailed analysis of the coding area by segmenting the skeleton, obtains a multi-segment coding skeleton, ensures the independent identification of each segment in the skeleton structure, facilitates the subsequent processing of the continuity of the characteristics of each segment, provides basic structural support for the length and angle difference analysis, enhances the segment recognition accuracy, compares the adjacent length differences of the multi-segment coding skeleton, generates the adjacent coding length differences by analyzing the length differences of adjacent skeleton segments, accurately evaluates the continuity of the coding skeleton in length, helps to identify the abnormal situation of length change, compares the adjacent angle differences of the multi-segment coding skeleton, obtains the adjacent coding angle differences by analyzing the angle change trend of adjacent skeleton segments, and ensures that the angle change of the coding stripes is within a reasonable range. Improve the judgment accuracy of the coding model on angle continuity, assist in detecting the occurrence of angle anomalies, conduct segmented continuity evaluation on the difference in length of adjacent codings and the difference in angle of adjacent codings, generate segmented continuity quality by comprehensively comparing the length and angle differences, accurately judge the overall continuity level of each segment skeleton, and provide accurate quantitative basis for the consistency judgment of coding quality. Based on the segmented continuity quality, map the continuity index of the basic coding skeleton, generate the skeleton continuity index through mapping, provide the overall consistency evaluation of the skeleton in the coding area, ensure the quantitative expression of the skeleton quality in terms of continuity, improve the accuracy and comprehensiveness of the coding quality evaluation, and provide the consistency index for the final coding quality judgment.

[0014] Preferably, step S4 comprises the following steps: Step S41: performing boundary positioning on the basic skeleton of the inkjet printer to obtain a set of stripe boundary points; Step S42: performing spacing traversal statistics on the stripe boundary point set to generate the stripe spacing distance; Step S43: Calculate the local discreteness of the coding basic skeleton based on the stripe spacing distance to obtain the coding gap discreteness; Step S44: extracting the mean of the inkjet coding gap discreteness to generate a discreteness mean; Step S45: Perform a uniformity assessment on the basic skeleton of the inkjet coding based on the preset uniformity quality reference value and the mean value of the dispersion to obtain the stripe uniformity quality.

[0015] The present invention obtains a stripe boundary point set by accurately identifying the boundary of the coding stripe, realizes the fine characterization of the stripe boundary position, provides accurate boundary data for subsequent spacing statistics, improves the resolution and overall contour clarity of the boundary point set, performs spacing traversal statistics on the stripe boundary point set, generates the stripe spacing distance by calculating the stripe boundary spacing, ensures the accurate quantification of the spacing of each stripe segment, provides specific data support for the subsequent discrete degree analysis, improves the measurement accuracy of the coding stripe spacing, calculates the local discrete degree of the coding basic skeleton based on the stripe spacing distance, obtains the coding gap discrete degree by localizing the spacing distance, and generates the discrete characteristic between each coding stripe. It ensures the regional distribution of discrete data, assists in analyzing the uniform distribution of coding stripes, increases the detail accuracy of stripes, extracts the mean of coding gap discreteness, generates the mean of discreteness by calculating the mean of each discrete value, ensures the overall overview of the discrete distribution of coding stripes, provides the mean reference for stripe uniformity quality evaluation, improves the accuracy of overall uniformity analysis, evaluates the uniformity of coding basic skeleton based on the preset uniformity quality reference value and the mean of discreteness, obtains the uniformity quality of stripes by comparing the standard reference with the mean of discreteness, ensures that the stripe distribution in the coding area meets the established uniformity quality standard, and realizes the quantitative determination of uniformity of coding quality.

[0016] Preferably, step S43 includes the following steps: Step S431: cutting the coding gap area of ​​the coding basic skeleton based on the stripe spacing distance to generate a coding gap area set; Step S432: determining the spacing extreme values ​​of the inkjet coding gap area set to obtain the maximum spacing value and the minimum spacing value; Step S433: Calculate the local discreteness of the coding gap area set according to the maximum spacing value and the minimum spacing value to generate a local gap discreteness set; Step S434: globally integrate the local gap discreteness set to obtain the coding gap discreteness.

[0017] The present invention generates a coding gap area set by identifying and cutting out the gap area between stripes, realizes the accurate distinction of the coding gap area, ensures the integrity and independence of the gap area, determines the spacing extreme value of the coding gap area set, obtains the maximum and minimum spacing values ​​of the gap spacing by calculating the maximum and minimum spacing values ​​in the gap area set, realizes the extreme value judgment of the spacing range of the gap area, calculates the local discreteness of the coding gap area set according to the maximum and minimum spacing values, calculates the discreteness in the local area by the guiding effect of the extreme spacing, generates a local gap discreteness set, ensures that the distribution regularity of the local gap spacing is quantified, realizes the local discretization analysis of the coding stripe gap, improves the local feature analysis depth of the gap area, performs global integration processing on the local gap discreteness set, obtains the coding gap discreteness by integrating and summarizing the local discreteness data, and realizes the transformation of local features to overall features.

[0018] Preferably, step S433 includes the following steps: The standard spacing difference is calculated for the maximum spacing value and the minimum spacing value to obtain the spacing standard deviation value; Calculate the fluctuation range of the maximum and minimum interval values ​​to generate a discrete allowable range; Based on the standard deviation of the spacing, an ideal interval is constructed for the inkjet coding gap area set to generate an ideal reference interval; According to the ideal reference interval, the relative position of the inkjet coding gap area set is located to obtain the regional deviation data; Identify the fluctuation interval of regional deviation data based on the discrete allowable range and generate the regional deviation fluctuation interval; The regional deviation fluctuation interval is evaluated for dispersion according to the ideal benchmark interval to generate a local gap dispersion set.

[0019] The present invention achieves standard deviation measurement of the set of coding gap areas by obtaining the standard deviation value of the spacing, ensures the quantification accuracy of the spacing data, provides the standard deviation basis of the coding gap area, calculates the fluctuation range of the maximum spacing value and the minimum spacing value, generates a discrete allowable range, and realizes standardized control of the fluctuation of the spacing of the coding gap area by clarifying the allowable range of discrete data, ensures the rationality and controllability of the discreteness evaluation, improves the accuracy of the fluctuation range evaluation, builds an ideal interval for the set of coding gap areas based on the standard deviation value of the spacing, generates an ideal reference interval, and provides a reference for the relative position of the coding gap by building the ideal interval, so that the distribution of discrete data is clearer and more orderly, the uniformity of data positioning is ensured, the benchmark comparability of the gap interval is strengthened, and the coding gap is evaluated according to the ideal reference interval. The relative position is located based on the regional set. By obtaining the regional deviation data, the position of each gap area relative to the ideal reference interval is made clearer, and the differentiated identification of the regional distribution is realized, which provides an accurate regional deviation basis for the subsequent discrete fluctuation analysis. The fluctuation interval of the regional deviation data is identified based on the discrete allowable range. By generating the regional deviation fluctuation interval, it is ensured that the fluctuation amplitude of the coding gap area meets the discrete control requirements, and the regional analysis of discrete data is realized. The hierarchical evaluation of the fluctuation distribution of the gap area is improved. The discreteness of the regional deviation fluctuation interval is evaluated according to the ideal reference interval. By generating the local gap discreteness set, it is ensured that the discreteness of the coding area is fully measured, and the local distribution characteristics of the coding gaps are accurately analyzed, which provides a basis for the discrete evaluation of the overall quality of coding.

[0020] Preferably, step S5 comprises the following steps: Step S51: performing weighted fusion on the stripe continuity quality and the stripe uniformity quality to obtain a fusion quality value; Step S52: performing readability quality deduction on the simulated inkjet coding model based on the fused quality value to generate a predicted readability quality value; Step S53: performing hierarchical processing on the predicted readable quality value to obtain a graded coding quality value; Step S54: mapping the quality level of the simulated inkjet printing model according to the graded inkjet printing quality value, and generating a quality assessment report.

[0021] The present invention ensures the balanced integration of various quality indicators through comprehensive weight calculation of stripe quality, provides a comprehensive quality evaluation method, accurately reflects the comprehensive state of inkjet coding quality, effectively improves the evaluation accuracy, and deduces the readability quality of the simulated inkjet coding model based on the fused quality value to generate a predicted readability quality value. By deducing and analyzing the fused quality value, it is possible to simulate the readability performance of inkjet coding under different conditions, predict the readability of actual inkjet coding, provide a basis for further quality control, enhance the predictability and reliability of inkjet coding quality, perform hierarchical processing on the predicted readability quality value, obtain a graded inkjet coding quality value, and by grading the predicted value, the inkjet coding quality can be divided into different grades according to preset standards, ensuring the accuracy and stratification of quality evaluation, making the inkjet coding quality level more intuitive and convenient for subsequent optimization and adjustment. According to the graded inkjet coding quality value, the simulated inkjet coding model is mapped to a quality grade, and a quality evaluation report is generated. By generating a detailed quality evaluation report, the visual presentation and scientific classification of inkjet coding quality are realized, a clear reference basis is provided for subsequent decision-making, and the transparency and systematicness of the inkjet coding quality detection process are enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic diagram of the steps of a method for detecting the quality of inkjet printing of an intelligent inkjet printer; Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart; Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG. The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0024] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0025] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0026] To achieve this, please refer to Figures 1 to 3 , a method for detecting the inkjet quality of an intelligent inkjet printer, comprising the following steps: Step S1: collecting an image of the coding area; performing double histogram equalization on the image of the coding area, and performing binary segmentation to obtain a binary image; Step S2: extracting the extreme connected domain of the binary image, and reconstructing the coding simulation to generate a simulation coding model; Step S3: splitting the stripe skeleton of the simulated inkjet coding model to obtain the basic skeleton of the inkjet coding; performing extension and coherence analysis on the basic skeleton of the inkjet coding to generate stripe continuity quality; Step S4: Based on the basic skeleton of inkjet printing, gap dispersion analysis is performed on the simulated inkjet printing model, and uniformity evaluation is performed to obtain stripe uniformity quality; Step S5: Deducing the readable quality of the simulated inkjet coding model according to the stripe continuity quality and the stripe uniformity quality, and performing quality grade mapping to generate a comprehensive quality assessment report.

[0027] The present invention adjusts the illumination and contrast of the image through double histogram equalization to ensure that the image detail information is clearly captured under different illumination conditions, and improves the contrast of the coding area in the image through binary segmentation, further highlights the edge contour of the character, and facilitates subsequent feature extraction. The binary image is subjected to extreme connected domain extraction to effectively remove noise interference, ensure that the character structure in the coding area is complete and clear, and restores the coding form through coding simulation reconstruction to generate a simulated coding model with a clear structure without affecting the content of the coding characters, providing image data with a complete structure for subsequent quality inspection, performing stripe skeleton splitting on the simulated coding model, retaining the main contour of the character, removing redundant information, and obtaining the basic coding skeleton. The analysis can accurately reflect the continuity of characters and generate the quality of stripe continuity, so that the continuity information between the coded characters can be clearly displayed, providing reliable data for evaluating the clarity and recognizability of the coding. The gap discreteness analysis of the simulated coding model is performed based on the basic skeleton of the coding. By evaluating the distribution of gaps, it is ensured that there are no obvious breaks and uniformity on the edges of the characters, and the presence of gaps and defects in the characters is accurately identified. The stripe uniformity quality is generated, and the readability quality of the simulated coding model is deduced according to the stripe continuity quality and stripe uniformity quality. The quality assessment results are visualized through quality grade mapping, and a comprehensive quality assessment report is generated to ensure the readability of the coded characters under different lighting, clarity and uniformity conditions, and provide quantifiable coding quality assessment results.

[0028] In the embodiment of the present invention, refer to Figure 1 , is a schematic diagram of the steps of a method for detecting the quality of inkjet printing of an intelligent inkjet printer of the present invention. In this example, the method for detecting the quality of inkjet printing of an intelligent inkjet printer includes the following steps: Step S1: collecting an image of the coding area; performing double histogram equalization on the image of the coding area, and performing binary segmentation to obtain a binary image; In this embodiment, a high-resolution camera is used to capture an image of the coding area. The image size should be no less than 1024×1024 pixels. During the capture, it should be ensured that the coding area is within a clear visible range, and the image should have sufficient illumination and contrast. Subsequently, the dual histogram equalization technology is used to process the brightness and contrast of the image. The original image is subjected to preliminary histogram equalization to increase the overall contrast of the image. Then, the histogram equalization of the local area is performed to optimize the local details. After the processing is completed, the image will become clearer and have a stronger detail contrast. Subsequently, the equalized image is binarized and converted into a black and white image by setting an appropriate threshold. The white part represents the coding area and the black part represents the background, thereby obtaining a binarized image. The threshold setting in the binarization process should ensure that the outline of the coding area in the image is clear and the background part is clearly separated from the coding area. The resulting binarized image will be used for subsequent analysis and processing.

[0029] Step S2: extracting the extreme connected domain of the binary image, and performing coding simulation reconstruction to generate a simulation coding model; In this embodiment, the white area in the binary image is subjected to extreme connected domain extraction. The connected domain refers to the area in the image with the same and adjacent pixel values. The 8-neighborhood connection rule is used to compare each pixel in the image with the 8 pixels around it to determine whether they belong to the same connected domain. All the connected domains are extracted, and then each extracted connected domain is further processed to remove small area noise points to ensure that the connected domain meets the size requirements of the coding area. After that, these connected domains are subjected to coding simulation reconstruction. By analyzing the shape, texture and other features of the connected domain, the distribution of inkjet points during the coding process is simulated to reconstruct a simulated coding model. The shape of the simulated coding model is as close as possible to the inkjet effect of the actual coding area. The model will serve as the basis for subsequent quality inspection.

[0030] Step S3: splitting the stripe skeleton of the simulated inkjet model to obtain the basic skeleton of the inkjet; performing extension and coherence analysis on the basic skeleton of the inkjet to generate stripe continuity quality; In this embodiment, the stripe skeleton of the simulated inkjet coding model is split, and the skeleton refers to a refined version of the inkjet coding area in the image. The edges of the inkjet coding area in the image are extracted to form thin lines. A skeletonization algorithm, such as the Zhang-Suen algorithm (Zhang Suen skeletonization algorithm), is used to remove redundant pixels in the skeleton and retain only the center line of the skeleton to obtain a basic inkjet coding skeleton. The basic inkjet coding skeleton represents the stripe structure of the inkjet coding, and its detailed features can reflect the quality of the inkjet coding. Then, an extension continuity analysis is performed on the obtained basic inkjet coding skeleton to analyze the continuity between each stripe in the skeleton. A path extension algorithm is used to detect breakpoints and connection points in the skeleton and calculate the continuity of each two adjacent skeleton parts. If there is a break or discontinuity, the extension algorithm will mark it as a low continuity area. The result of the extension continuity analysis will generate the stripe continuity quality. The higher the evaluation value of the stripe continuity quality, the better the ductility of the inkjet coding stripes and the higher the inkjet coding quality.

[0031] Step S4: Based on the basic skeleton of inkjet printing, gap dispersion analysis is performed on the simulated inkjet printing model, and uniformity evaluation is performed to obtain stripe uniformity quality; In this embodiment, a gap discreteness analysis is performed based on the basic skeleton of the inkjet coding. The gap discreteness refers to the change in the spacing between the inkjet coding stripes. A statistical method is used to calculate the gap distance between each stripe in the skeleton, and the standard deviation of these distances is calculated. The smaller the standard deviation, the more uniform the spacing of the inkjet coding stripes. The obtained gap discreteness data is then evaluated for uniformity. By comparing with a preset ideal spacing range, the uniformity of the inkjet coding stripes is evaluated. If the gap discreteness exceeds the predetermined range, it indicates that the uniformity of the inkjet coding is poor, which will result in poor readability of the inkjet coding or affect the appearance quality. Finally, an evaluation value of the stripe uniformity quality is obtained. The higher the value, the better the uniformity of the inkjet coding.

[0032] Step S5: Deducing the readable quality of the simulated inkjet coding model according to the stripe continuity quality and the stripe uniformity quality, and performing quality grade mapping to generate a comprehensive quality assessment report.

[0033] In this embodiment, the readability quality of the simulated inkjet coding model is deduced based on the stripe continuity quality and the stripe uniformity quality. The two quality indicators are integrated using a weighted average algorithm to generate a comprehensive readability quality deduction value. The higher the deduction value, the better the readability of the inkjet coding. Then, quality grade mapping is performed based on the deduced value, and multiple grade ranges are set, such as excellent, good, fair, unqualified, etc. The final quality grade of the inkjet coding is determined based on the deduced value. Finally, a comprehensive quality assessment report is generated, which includes information such as stripe continuity quality, stripe uniformity quality, readability quality deduction results and quality grade. The report can provide detailed quality feedback for the inkjet printer operator.

[0034] Preferably, step S1 comprises the following steps: Step S11: collecting an image of the coding area; performing global histogram equalization processing on the image of the coding area to obtain an enhanced coding image; Step S12: performing non-enhanced area detection on the enhanced inkjet coding image to obtain a non-uniform area; Step S13: performing local adaptive equalization processing on the non-uniform area to generate a double equalized image; Step S14: performing maximum inter-class variance analysis on the double equalized image to obtain a cutting variance threshold parameter; Step S15: performing binarization segmentation on the double equalized image based on the cutting variance threshold parameter to obtain a binarized image.

[0035] In this embodiment, a high-resolution image of the coding area is collected by an industrial camera to ensure the clarity of details. The collected image of the coding area is subjected to global histogram equalization. The equalization process uses the Histogram Equalization algorithm to enhance the image contrast by stretching and redistributing pixel values, so that the light and dark contrast of the coding area is more significant. The global histogram equalization is suitable for scenes where the overall brightness of the image is unevenly distributed. The enhanced coding image is more uniform in color scale. The uneven area in the image is detected. The uneven area detection adopts a local contrast analysis method. The image is traversed pixel by pixel through a sliding window (such as 9×9 or 11×11 pixel size) to detect the pixel contrast difference in each area. If the pixel brightness difference in a certain area is lower than a preset threshold, the area is marked as an uneven area. The uneven area mainly includes parts of the image with insufficient contrast or too low brightness. These areas are difficult to accurately cut during the binary segmentation process, and therefore must be marked separately. For the detected uneven area, local adaptive histogram equalization is performed to enhance the local contrast of the image. Local adaptive equalization (Adaptive Histogram Equalization, AHE) divides the image into multiple small windows and performs histogram equalization on each window area to achieve uniform distribution of local brightness of the image. In order to prevent artifacts or noise caused by over-equalization, the contrast limit parameter (such as 0.03 to 0.05) is set to control the processing effect. The generated double-equalized image has a high contrast level in both the overall and local areas. The maximum inter-class variance analysis is performed on the double-equalized image to obtain the cutting variance threshold parameter, the maximum inter-class variance (Otsu's Method, Otsu's method is a threshold segmentation method based on inter-class variance. It calculates the inter-class variance between the foreground and background in the image, and selects the threshold that maximizes the inter-class variance as the basis for segmentation. In the operation, the algorithm is used to divide the grayscale values ​​of the double equalized image into two categories, and the intra-class variance and inter-class variance are calculated respectively. Then, the grayscale value that maximizes the inter-class variance is selected as the optimal segmentation threshold. Based on the cutting variance threshold parameter, the double equalized image is binarized and segmented. The binary segmentation divides the pixel values ​​in the image into two categories through the set threshold. The foreground is displayed as black and the background is white. The binarization process uses the Thresholding algorithm to set pixels greater than or equal to the cutting variance threshold to black, and pixels less than this value to white, thereby obtaining a clear binary image.

[0036] Preferably, step S2 comprises the following steps: Step S21: Perform eight-neighborhood channel statistics on the binary image to obtain a candidate connected domain; Step S22: Tracing the boundary contour of the candidate connected domain to generate a connected domain contour point set; Step S23: identifying extreme points of the connected domain contour point set to obtain key contour points of the connected domain; Step S24: performing feature topological relationship analysis on key contour points of the connected domain to generate a contour point topological structure; Step S25: reconstructing the key contour points of the connected domain based on the contour point topological structure to obtain the key contour for inkjet printing; Step S26: Fill the key outline of the inkjet coding with inkjet texture according to the binary image to generate a simulated inkjet coding model.

[0037] As an example of the present invention, see Figure 2 In this example, step S2 includes: Step S21: Perform eight-neighborhood channel statistics on the binary image to obtain a candidate connected domain; In this embodiment, an eight-neighborhood channel statistics algorithm is used to identify candidate connected domains in an image. The eight-neighborhood channel statistics is a technical method for determining the neighborhood connection relationship of each pixel. The method determines whether each pixel belongs to the same connected area by checking the color consistency of its neighboring pixels in eight directions (left, right, top, bottom, top left, top right, bottom left, and bottom right) around it. The specific operation is to scan line by line starting from the upper left corner of the image, group each pixel according to the eight-neighborhood rule, and if adjacent pixels belong to the same connected block, they are marked as the same connected domain, and their pixel coordinates are recorded. After the scan is completed, all candidate connected domains and their coordinate information in the coding area are obtained, and each candidate connected domain is marked for subsequent processing.

[0038] Step S22: Tracing the boundary contour of the candidate connected domain to generate a connected domain contour point set; In this embodiment, a boundary contour tracing operation is performed on the obtained candidate connected domain to generate a contour point set of the candidate connected domain. The contour tracing adopts the chain code method (Chain Code), which tracks the outer boundary pixels of each connected domain in turn, records the position information of the boundary points to generate a complete contour point set, starts from the starting point of the connected domain (the upper left corner pixel), and tracks pixel by pixel in a clockwise direction. If a boundary switch is encountered, the coordinates of the point are recorded and stored in the contour point set until the closed-loop contour is completed by returning to the starting point. The contour point set finally generated contains the boundary contours of all candidate connected domains.

[0039] Step S23: identifying extreme points of the connected domain contour point set to obtain key contour points of the connected domain; In this embodiment, extreme point recognition is performed on the generated connected domain contour point set to obtain the key contour points of the connected domain. Extreme point recognition refers to obtaining the vertex information and boundary endpoints of the contour by calculating the maximum and minimum values ​​of each coordinate point in the contour point set in a specific direction, extracting the extreme points of the contour point set in the horizontal and vertical directions, and then calculating the corner points and marking the corner points where the contour changes drastically. These corner points and endpoints are used as the key contour points of the connected domain, and the obtained extreme point data is stored to form a set of key contour points of the connected domain.

[0040] Step S24: performing feature topological relationship analysis on key contour points of the connected domain to generate a contour point topological structure; In this embodiment, a topological relationship analysis is performed on a set of key contour points in a connected domain to generate a topological structure of the contour points. The topological relationship analysis refers to analyzing the spatial connection relationship and relative position between each key point and adjacent points. A connection structure between contour points is established through a Delaunay triangulation mesh. Contour points and adjacent points are connected to form triangles to avoid intersections and ensure the integrity of the topological structure. The generated topological structure information includes the adjacent point sequence and geometric angle relationship of each key point in the connected domain. This information is used as a basic structure in contour reconstruction.

[0041] Step S25: reconstructing the key contour points of the connected domain based on the contour point topological structure to obtain the key contour for inkjet printing; In this embodiment, the key contour points of the connected domain are reconstructed based on the topological structure of the contour points. The contour reconstruction forms a complete inkjet contour structure by redrawing each edge line marked in the topological structure into a coherent curve. The contour reconstruction adopts the bicubic spline interpolation method to make the curves smoothly connected. In the specific operation, the key contour points are curve fitted according to the edge connection order of the topological structure, each edge line is interpolated into a smooth curve, and the global contour reconstruction is completed according to the arrangement of the contour points to obtain a clear and complete inkjet key contour.

[0042] Step S26: Fill the key outline of the inkjet coding with inkjet texture according to the binary image to generate a simulated inkjet coding model.

[0043] In this embodiment, texture filling is performed on the key contour of the coding according to the binary image to generate a simulated coding model. During the texture filling process, the pixel values ​​of the coding area of ​​the binary image are used as texture information, and a scan line-based filling algorithm is used to fill the area within the key contour of the coding pixel by pixel. The scan line filling method starts from the top of the coding contour and fills row by row to the bottom. The pixels within the boundary of each row are filled with the coding color to ensure seamless coverage of the texture within the contour. The finally generated simulated coding model contains the real coding texture and edge information for subsequent detection and analysis.

[0044] Preferably, step S3 comprises the following steps: Step S31: positioning the contour branch points of the simulation inkjet model to obtain inkjet branch position information; Step S32: splitting the simulated inkjet model into skeletons based on the inkjet branch position information to generate an inkjet basic skeleton; Step S33: monitoring the extension breakpoints of the coding basic skeleton to obtain the skeleton breakpoint positions; Step S34: performing a coherence analysis on the basic skeleton of the coding based on the skeleton breakpoint positions to generate a skeleton continuity index; Step S35: mapping the skeleton continuity index to a quality level according to a preset continuity quality reference value to generate a stripe continuity quality.

[0045] As an example of the present invention, see Figure 3 In this example, step S3 includes: Step S31: positioning the contour branch points of the simulation inkjet model to obtain inkjet branch position information; In this embodiment, contour branch point positioning is performed on the simulated inkjet coding model to identify branch node position information during the inkjet coding process. Branch node position information refers to position information where there are bifurcations or corners in the inkjet coding lines. The contour of the inkjet coding model is iteratively reduced multiple times through a thinning algorithm to retain its skeleton structure and identify branches in the skeleton. A branch point refers to a key node with more than two connection points in the skeleton structure. To ensure accuracy, a structured element analysis is first performed to confirm whether the branch points of the skeleton contour actually exist and to exclude noise points. After completion, all branch nodes are stored in the inkjet coding branch position information set and marked.

[0046] Step S32: splitting the simulated inkjet model into skeletons based on the inkjet branch position information to generate an inkjet basic skeleton; In this embodiment, based on the inkjet branch position information, the simulated inkjet model is skeletonized to generate a basic inkjet skeleton. The skeleton splitting adopts a cutting algorithm (Cutting Algorithm), and the branch nodes are used as cutting points to split the overall skeleton into multiple independent skeleton fragments. These fragments represent the basic structure of the inkjet model. The characteristics of each fragment are analyzed segment by segment to confirm its independence. At the same time, connection indexes are set at the beginning and end of each fragment for subsequent processing. During the splitting process, each skeleton fragment retains the contour characteristics of the original inkjet.

[0047] Step S33: monitoring the extension breakpoints of the coding basic skeleton to obtain the skeleton breakpoint positions; In this embodiment, the generated basic skeleton for inkjet coding is subjected to extension breakpoint monitoring to determine potential breakpoint positions in the skeleton. The extension breakpoint monitoring gradually tracks the endpoints of each skeleton segment to find out whether there is a disconnection or discontinuity in the extension direction. In the specific operation, each skeleton segment is tracked pixel by pixel from the starting point along the contour direction to detect whether the distance between adjacent pixels exceeds the preset breakpoint threshold. If an over-distance phenomenon occurs, the location is recorded as the breakpoint location and stored in the skeleton breakpoint location set. The existence of a breakpoint will indicate a coding quality problem. By recording the breakpoint location, a data basis is provided for the subsequent continuity analysis.

[0048] Step S34: performing a coherence analysis on the basic skeleton of the coding based on the skeleton breakpoint positions to generate a skeleton continuity index; In this embodiment, a continuity analysis is performed on the coding basic skeleton according to the positions of the skeleton breakpoints to generate a skeleton continuity index. The continuity analysis determines the integrity of the overall skeleton by calculating the average continuous length of each skeleton segment and the breakpoint interval distance. The specific method is to count the number and position of continuous points in each skeleton segment, and to form a continuity index by counting the distance between each breakpoint in the skeleton and performing a weighted average calculation. The continuity index represents the degree of continuity of the skeleton. The fewer the breakpoints and the shorter the intervals, the more complete the skeleton. The analysis results form a set of skeleton continuity indicators.

[0049] Step S35: mapping the skeleton continuity index to a quality level according to a preset continuity quality reference value to generate a stripe continuity quality.

[0050] In this embodiment, the skeleton continuity index is mapped to a quality grade according to a preset continuity quality reference value to generate a stripe continuity quality. The skeleton continuity index is compared with the preset reference value, and the index is divided into different grades such as "excellent", "good", "medium" and "poor" according to the set quality grade standard. The specific operation is to determine the corresponding quality grade according to the deviation ratio between the specific value of the skeleton continuity index and the reference value. For example, if the continuity index is within the reference value range, the quality grade is judged to be "excellent". As the deviation increases, the grade decreases successively. The stripe continuity quality grade finally generated will be used for the final detection and evaluation of the coding quality.

[0051] Preferably, step S34 includes the following steps: The basic coding skeleton is segmented based on the skeleton breakpoint positions to obtain a multi-segment coding skeleton; Compare the adjacent length differences of multiple sections of inkjet skeletons to generate the adjacent inkjet length differences; Compare the adjacent angle differences of multiple sections of inkjet skeletons to obtain the adjacent inkjet angle differences; Perform segment continuity evaluation on the difference in adjacent code lengths and adjacent code angles to generate segment continuity quality; Based on the segmented continuity quality, the coding basic skeleton is mapped to the continuity index to generate the skeleton continuity index.

[0052] In this embodiment, the skeleton breakpoint position is used to segment the basic skeleton of the inkjet coding, and the skeleton fragments between each breakpoint are used as independent skeleton segments. Each breakpoint is positioned on the basic skeleton of the inkjet coding in turn, and the adjacent breakpoints are used as the starting and ending points of the segmentation. If the length of the skeleton segment is too small, the segment is skipped to avoid the accumulation of errors in the analysis. Through this segmentation operation, several independent multi-segment inkjet coding skeletons are obtained. The structure of each skeleton segment remains consistent to ensure the equivalence of each segment in subsequent analysis. The actual pixel length of each skeleton segment is measured in turn, the total number of pixels of each segment is recorded, and the lengths of adjacent segments are subtracted. The length difference of adjacent codes is obtained. The length difference indicates the length change of the code skeleton between the segments to judge the uniformity of the code. Finally, each length difference result is stored in the adjacent code length difference set. The size of the length difference value is used to evaluate the overall uniformity and length distribution characteristics of the code. In the angle difference comparison, the direction angle of the multi-segment code skeleton is calculated segment by segment. The starting point and the end point of each segment are used as the reference for angle calculation to obtain the direction vector of each segment skeleton. The angle between the direction vectors of adjacent segments is calculated. The angle is the angle difference between the two segments. If the angle difference between adjacent segments is too large, it indicates that the code is uneven. There is a problem with the continuity of the code. The angle difference values ​​of each adjacent segment are recorded in sequence to form a set of adjacent coding angle differences. The closer the values ​​in the angle difference set are, the more consistent the direction of the coding skeleton is. The larger the angle difference is, the discontinuity exists in the coding process. The continuity of the coding is comprehensively evaluated based on the length difference and angle difference. The length difference and angle difference of each skeleton segment are used as the consistency judgment criteria. The length difference and angle difference are weighted to obtain the consistency score between each skeleton segment. If the score is within the consistency standard range, the segment is marked as "coherent", otherwise it is marked as " Incoherence", and finally summarize the continuity marks of each skeleton segment to generate the overall segmented continuity quality result, which shows the uniformity and continuity characteristics of the coding. The segmented continuity quality result is mapped and analyzed with the continuity information of each segment of the basic coding skeleton to statistically score the continuity of each coherent segment. Combined with the preset continuity standard, the overall continuity score of the skeleton is calculated to complete the mapping of the continuity index of the coding skeleton. The higher the score, the more continuous the coding. The generated skeleton continuity index will be used as the basis for judging the coding quality. This index is used to quantify the coding continuity and for the comprehensive evaluation of the subsequent coding quality.

[0053] Preferably, step S4 comprises the following steps: Step S41: performing boundary positioning on the basic skeleton of the inkjet printer to obtain a set of stripe boundary points; Step S42: performing spacing traversal statistics on the stripe boundary point set to generate the stripe spacing distance; Step S43: Calculate the local discreteness of the coding basic skeleton based on the stripe spacing distance to obtain the coding gap discreteness; Step S44: extracting the mean of the inkjet coding gap discreteness to generate a discreteness mean; Step S45: Perform a uniformity assessment on the basic skeleton of the inkjet coding based on the preset uniformity quality reference value and the mean value of the dispersion to obtain the stripe uniformity quality.

[0054] In this embodiment, each pixel point of the coding skeleton is processed based on the edge detection algorithm, and the edge pixel points of each section of the coding are identified in turn using the contour detection function. All the identified edge pixel points will be stored in the stripe boundary point set, which contains the specific position of each stripe in space, so as to ensure that the boundary point set has sufficient accuracy in the subsequent spacing statistics. According to the arrangement order of the stripes, the coordinate positions of adjacent boundary points are extracted from the boundary point set one by one, and the Euclidean distance between each adjacent boundary point is calculated and stored in the stripe interval distance list. The value of the stripe interval distance list is used to represent the interval regularity of the coding stripes and the uniformity of the coding. The statistics of each interval distance are helpful for the subsequent discrete calculation. The values ​​in the stripe interval distance list are divided into several areas, and then the stripe interval values ​​in each area are subjected to standard deviation. Deviation) calculation is performed to obtain the discreteness value of each area, and these discreteness values ​​are stored one by one in the inkjet gap discreteness list. The size of the inkjet gap discreteness reflects whether the interval between the inkjet stripes is consistent. A larger discreteness value indicates that the interval between the stripes is larger, and a smaller discreteness value indicates that the stripe interval is more uniform. The discreteness values ​​in the inkjet gap discreteness list are accumulated, and the accumulated result is divided by the number of discreteness values ​​to obtain the discreteness mean. The discreteness mean is used to quantify the uniformity characteristics of the overall inkjet printing. A lower mean value indicates that the stripe interval of the inkjet printing is more consistent, and a high discreteness value indicates that the stripe interval is more uniform. The mean value indicates that there is obvious non-uniformity in the stripe spacing. The mean value is finally stored in the dispersion mean variable for subsequent uniformity evaluation. The dispersion mean value is compared with the preset uniformity quality reference value. The uniformity quality reference value is a threshold value set based on the inkjet coding process standard, which is used to judge the uniformity of the inkjet coding stripes. If the dispersion mean value is less than or equal to the uniformity quality reference value, the inkjet coding uniformity quality is judged to be qualified, otherwise it is judged to be unqualified. The evaluation result is stored in the stripe uniformity quality variable. The uniformity quality value will be used as the final judgment basis for the inkjet coding quality to measure the overall uniformity of the inkjet coding process.

[0055] Preferably, step S43 includes the following steps: Step S431: cutting the coding gap area of ​​the coding basic skeleton based on the stripe spacing distance to generate a coding gap area set; Step S432: determining the spacing extreme values ​​of the inkjet coding gap area set to obtain the maximum spacing value and the minimum spacing value; Step S433: Calculate the local discreteness of the coding gap area set according to the maximum spacing value and the minimum spacing value to generate a local gap discreteness set; Step S434: globally integrate the local gap discreteness set to obtain the coding gap discreteness.

[0056] In this embodiment, the stripe spacing distance list is combined with the basic skeleton of the coding to calibrate the gap position between the coding stripes. By analyzing the spacing greater than the set threshold in the stripe spacing distance list, the gap area of ​​the coding stripes is separated from the basic skeleton one by one. Each area that meets the gap condition will be cut and stored in the coding gap area set to form a gap set containing the position and boundary of each gap area. After the gap area is divided, the gap area set will be used for subsequent extreme value and discreteness analysis operations. Each gap area in the coding gap area set is detected one by one, and the width of each gap area is measured. The measured values ​​are stored in the gap spacing list in turn, and the extreme value function (Extreme Value) is applied to the gap spacing list. Function), identify the maximum and minimum values, and store them as the maximum and minimum values ​​of gap spacing, respectively. The extreme value of gap spacing will be used as the basic parameter for local discreteness calculation to evaluate the gap variation range of stripes during the coding process. The difference analysis is performed on each area in the gap area set according to the maximum and minimum values. The difference between the relative extreme values ​​of the spacing of each segment in the gap area is calculated, and the variance function is used to perform discreteness operation to obtain the local discreteness value of the area, and all discreteness values ​​are stored as a local gap discreteness set. The local gap discreteness set is used to represent the spacing variation of each gap area to reflect the local uniformity difference between the coding stripes. The weighted average operation is performed on each discreteness value in the local gap discreteness set, and the calculation result will be defined as the coding gap discreteness. Through the total discreteness statistics, the coding gap discreteness value will be used to quantify the overall uniformity of each gap during the coding process. Finally, the coding gap discreteness, as the core indicator for evaluating the uniformity of coding, will be used to determine the results of coding quality inspection.

[0057] Preferably, step S433 includes the following steps: The standard spacing difference is calculated for the maximum spacing value and the minimum spacing value to obtain the spacing standard deviation value; Calculate the fluctuation range of the maximum and minimum interval values ​​to generate a discrete allowable range; Based on the standard deviation of the spacing, an ideal interval is constructed for the inkjet coding gap area set to generate an ideal reference interval; According to the ideal reference interval, the relative position of the inkjet coding gap area set is located to obtain the regional deviation data; Identify the fluctuation interval of regional deviation data based on the discrete allowable range and generate the regional deviation fluctuation interval; The regional deviation fluctuation interval is evaluated for dispersion according to the ideal benchmark interval to generate a local gap dispersion set.

[0058] In this embodiment, the maximum spacing value and the minimum spacing value are used as input data, and the standard deviation formula is applied, that is, the difference between the two is divided by the total number to obtain the sample standard deviation, thereby generating the spacing standard deviation value, and the spacing standard deviation value is used as an important indicator to measure the volatility of the coding spacing, which is used to reflect the discrete degree of the coding spacing. This value will provide a data basis for the subsequent construction of the ideal interval and serve as a reference for the fluctuation of the spacing. The allowable range between the maximum spacing value and the minimum spacing value is calculated using the discreteness analysis method. First, the allowable fluctuation coefficient is defined, and based on the coefficient, the difference between the maximum spacing value and the minimum spacing value is multiplied by the allowable coefficient to obtain the value of the spacing standard deviation. Generate a discrete allowable range, and use the discrete allowable range as the deviation limit for defining the spacing of the coding stripes for subsequent deviation evaluation to ensure the compliance of the spacing of the coding stripes. This range provides a reference standard for discrete detection of the coding gap area. By calculating and analyzing the standard deviation of the spacing, an ideal reference interval is constructed. The ideal reference interval is defined by the ideal spacing range defined by the standard deviation as the target interval of the coding stripe distribution. By introducing an ideal spacing range calibration point in each gap area of ​​the coding gap area set, it is convenient for locating and fluctuating the subsequent deviation. The ideal reference interval is used as the reference spacing set for evaluation. The fluctuation deviation of the actual inkjet coding spacing is estimated. The position of each inkjet coding gap area is compared with the reference interval by the preset standard based on the ideal reference interval, so as to measure the deviation amplitude of each gap area. The regional deviation calculation formula is used to count the distance difference between each inkjet coding area in the gap area set and the ideal interval and store it as deviation data. The generated regional deviation data is used for the subsequent identification of the fluctuation interval to determine the actual discrete deviation degree of the inkjet coding spacing. Based on the upper and lower limits of the discrete allowable range, the regional deviation data is compared with the allowable range one by one to determine whether the deviation of each area exceeds the specified discrete allowable range. The allowable range is defined, and the deviation data exceeding the allowable range is marked as the fluctuation abnormal interval. The regional deviation fluctuation interval is generated by recording the deviation amplitude of the abnormal interval. The regional deviation fluctuation interval is used as the deviation representation of the uniformity quality of the coding, so as to evaluate the stability of the coding process. The data points in the regional deviation fluctuation interval are compared with the ideal reference interval. The data distribution in the fluctuation interval is analyzed, and the deviation dispersion value of each area is calculated using the dispersion evaluation formula. The dispersion results of each area are successively integrated into the local gap dispersion set, and the local spacing fluctuation of the coding gap area is reflected by the local gap dispersion set.

[0059] Preferably, step S5 comprises the following steps: Step S51: performing weighted fusion on the stripe continuity quality and the stripe uniformity quality to obtain a fusion quality value; Step S52: performing readability quality deduction on the simulated inkjet coding model based on the fused quality value to generate a predicted readability quality value; Step S53: performing hierarchical processing on the predicted readable quality value to obtain a graded coding quality value; Step S54: mapping the quality level of the simulated inkjet printing model according to the graded inkjet printing quality value, and generating a quality assessment report.

[0060] In this embodiment, the stripe continuity quality value and the stripe uniformity quality value are obtained. The stripe continuity quality value indicates the continuity between the coding stripes, while the stripe uniformity quality value reflects the uniformity of the coding stripes. The two are important criteria for evaluating the quality of coding. They are fused by setting a weight coefficient, and the average weight method is adopted, that is, the weight of each quality value is set to 0.5. The fused quality value is calculated by weighted summation. The fused quality value represents the comprehensive quality level of the coding stripes and serves as the basis for subsequent quality evaluation. The fused quality value is used as input data and a simulation coding model is used. The model is based on previous coding data and deduces the readability of the coding by simulating various quality changes that occur during the coding process. In this process, the simulation model simulates the actual readability of the coding stripes according to the fused quality value and the simulation rules, and generates a predicted readable quality value. The predicted readable quality value reflects the readability of the coding stripes in actual applications. The predicted value will be used in the next step. The quality grading of the coding is provided with a reference basis, and several coding quality grades are defined, such as excellent, good, general, unqualified, etc. The limit value of each grade is set by the historical coding data and quality standards. The predicted readable quality value is then compared with these standards. If the predicted value falls within a certain range, the coding quality value is classified into the corresponding grade. Through this process, a graded coding quality value is generated, and the quality of each coding stripe will be assigned to the corresponding quality grade to facilitate subsequent quality evaluation and report generation. By using the graded coding quality value, a mapping relationship is established to associate the quality grade of each coding stripe with the specific area or coding point in the simulated coding model. Through the mapping, the quality status of the coding area is identified, and finally a quality assessment report is generated according to the mapping results. The report lists the quality grade and related indicators of each coding area in detail, and summarizes and evaluates the overall coding quality. The report can be used as a basis for debugging or repairing coding equipment, and also provides data support for quality control.

[0061] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0062] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for detecting the inkjet quality of an intelligent inkjet printer, characterized in that: The following steps are involved: Step S1: Collecting the image of the coding area; Perform double histogram equalization on the coding area image and perform binary segmentation to obtain a binary image; Step S2: extracting the extreme connected domain of the binary image, and reconstructing the coding simulation to generate a simulation coding model; Step S3: splitting the stripe skeleton of the simulated inkjet coding model to obtain the basic skeleton of the inkjet coding; performing extension and coherence analysis on the basic skeleton of the inkjet coding to generate stripe continuity quality; Step S4: Based on the basic skeleton of inkjet printing, gap dispersion analysis is performed on the simulated inkjet printing model, and uniformity evaluation is performed to obtain stripe uniformity quality; Step S5: Deducing the readable quality of the simulated inkjet coding model according to the stripe continuity quality and the stripe uniformity quality, and performing quality grade mapping to generate a comprehensive quality assessment report.

2. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: collecting the image of the coding area; performing global histogram equalization processing on the image of the coding area to obtain an enhanced coding image; Step S12: performing non-enhanced area detection on the enhanced inkjet coding image to obtain a non-uniform area; Step S13: performing local adaptive equalization processing on the non-uniform area to generate a double equalized image; Step S14: performing maximum inter-class variance analysis on the double equalized image to obtain a cutting variance threshold parameter; Step S15: performing binarization segmentation on the double equalized image based on the cutting variance threshold parameter to obtain a binarized image.

3. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: Perform eight-neighborhood channel statistics on the binary image to obtain a candidate connected domain; Step S22: Tracing the boundary contour of the candidate connected domain to generate a connected domain contour point set; Step S23: identifying extreme points of the connected domain contour point set to obtain key contour points of the connected domain; Step S24: performing feature topological relationship analysis on key contour points of the connected domain to generate a contour point topological structure; Step S25: reconstructing the key contour points of the connected domain based on the contour point topological structure to obtain the key contour for inkjet printing; Step S26: Fill the key outline of the inkjet coding with inkjet texture according to the binary image to generate a simulated inkjet coding model.

4. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: positioning the contour branch points of the simulation inkjet model to obtain inkjet branch position information; Step S32: splitting the simulated inkjet model into skeletons based on the inkjet branch position information to generate an inkjet basic skeleton; Step S33: monitoring the extension breakpoints of the coding basic skeleton to obtain the skeleton breakpoint positions; Step S34: performing a coherence analysis on the basic skeleton of the coding based on the skeleton breakpoint positions to generate a skeleton continuity index; Step S35: mapping the skeleton continuity index to a quality level according to a preset continuity quality reference value to generate a stripe continuity quality.

5. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 4 is characterized in that: Step S34 includes the following steps: The basic coding skeleton is segmented based on the skeleton breakpoint positions to obtain a multi-segment coding skeleton; Compare the adjacent length differences of multiple sections of inkjet skeletons to generate the adjacent inkjet length differences; Compare the adjacent angle differences of multiple sections of inkjet skeletons to obtain the adjacent inkjet angle differences; Perform segment continuity evaluation on the difference in adjacent code lengths and adjacent code angles to generate segment continuity quality; Based on the segmented continuity quality, the coding basic skeleton is mapped to the continuity index to generate the skeleton continuity index.

6. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: performing boundary positioning on the basic skeleton of the inkjet printer to obtain a set of stripe boundary points; Step S42: performing spacing traversal statistics on the stripe boundary point set to generate the stripe spacing distance; Step S43: Calculate the local discreteness of the coding basic skeleton based on the stripe spacing distance to obtain the coding gap discreteness; Step S44: extracting the mean of the inkjet coding gap discreteness to generate a discreteness mean; Step S45: Perform a uniformity assessment on the basic skeleton of the inkjet coding based on the preset uniformity quality reference value and the mean value of the dispersion to obtain the stripe uniformity quality.

7. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 6, characterized in that: Step S43 includes the following steps: Step S431: cutting the coding gap area of ​​the coding basic skeleton based on the stripe spacing distance to generate a coding gap area set; Step S432: determining the spacing extreme values ​​of the inkjet coding gap area set to obtain the maximum spacing value and the minimum spacing value; Step S433: Calculate the local discreteness of the coding gap area set according to the maximum spacing value and the minimum spacing value to generate a local gap discreteness set; Step S434: globally integrate the local gap discreteness set to obtain the coding gap discreteness.

8. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 7, characterized in that: Step S433 includes the following steps: The standard spacing difference is calculated for the maximum spacing value and the minimum spacing value to obtain the spacing standard deviation value; Calculate the fluctuation range of the maximum and minimum interval values ​​to generate a discrete allowable range; Based on the standard deviation of the spacing, an ideal interval is constructed for the inkjet coding gap area set to generate an ideal reference interval; According to the ideal reference interval, the relative position of the inkjet coding gap area set is located to obtain the regional deviation data; Identify the fluctuation interval of regional deviation data based on the discrete allowable range and generate the regional deviation fluctuation interval; The regional deviation fluctuation interval is evaluated for dispersion according to the ideal benchmark interval to generate a local gap dispersion set.

9. The method for detecting the inkjet quality of an intelligent inkjet printer according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: performing weighted fusion on the stripe continuity quality and the stripe uniformity quality to obtain a fusion quality value; Step S52: performing readability quality deduction on the simulated inkjet coding model based on the fused quality value to generate a predicted readability quality value; Step S53: performing hierarchical processing on the predicted readable quality value to obtain a graded coding quality value; Step S54: mapping the quality level of the simulated inkjet printing model according to the graded inkjet printing quality value, and generating a quality assessment report.

Images