A fine observation system for printed still images with a real-time self-check function

Abstract

The present invention relates to the technical field of image processing, and particularly relates to a fine observation system for printed still images with a real-time self-checking function. An image acquisition module is used to acquire an image of a printed product during the printing process; a defect screening module is used to perform local segmentation on the image of the printed product; perform multiple rounds of downsampling iteration on different local regions, analyze the local information loss of each local region during each downsampling iteration, and screen out high-printing information regions; according to the edge lines and regional color differences in the high-printing information regions, screen out printed defect regions from the high-printing information regions; a defect determination module is used to compare the printed defect regions with a printing template to identify the printed defects in the printed defect regions. The present invention reduces the amount of calculation in the defect identification process and improves the accuracy of defect identification.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a fine observation system for printed still images with a real-time self-check function. Background Art

[0002] In the application of traditional printing technology, for example, when performing overall printing on paper packaging materials, common problems such as uneven printing, large indentation, low yield, paper path loss, complex dot structure, and low reduction degree of gradient color dots are often encountered. These problems limit the clarity and fineness of printed patterns, make personalized customization difficult, result in high costs, relatively low production efficiency, and limit market competitiveness. Moreover, during printing, due to environmental factors and equipment reasons, printing defects such as misprinting and color difference may occur during the printing process. When completely checking the patterns of the entire printed product, the time cost will inevitably increase due to the large amount of calculation, affecting the efficiency of normal printing production. Currently, the existing fine observation of printed still images usually uses deep learning to directly identify the images of printed products by a trained model. However, due to the complex model, the processing speed may be slow on a high-speed production line, making it difficult to achieve self-checking of printed products at a small time cost. In addition, the accuracy of identifying some small defects is relatively low. For example, for tiny defects, it is easy to miss detection. Summary of the Invention

[0003] In order to solve the technical problem of relatively low accuracy in identifying small defects in printed product images, the purpose of the present invention is to provide a fine observation system for printed still images with a real-time self-check function, and the specific technical solutions adopted are as follows:

[0004] In a first aspect, an embodiment of the present invention provides a fine observation system for printed still images with a real-time self-check function, and the system includes the following modules:

[0005] An image acquisition module, configured to acquire an image of a printed product during the printing process;

[0006] A defect screening module, configured to perform local segmentation on the image of the printed product; perform multiple rounds of downsampling iteration on different local regions, analyze the local information loss of each local region during each downsampling iteration, screen out high-printing information regions; and screen out printing defect regions from the high-printing information regions according to the edge lines and regional color differences in the high-printing information regions;

[0007] A defect determination module, configured to compare the printing defect regions with a printing template and identify the printing defects in the printing defect regions.

[0008] Further, the method for screening out high-printing-information regions by analyzing the local information loss of each local region in each downsampling iteration includes:

[0009] Calculate the structural similarity of the region before and after downsampling for each local region, perform a negative correlation mapping process on the structural similarity to obtain the local information loss of each local region; for each local region, when the local information loss of the local region exceeds the preset normal loss range, stop the downsampling iteration.

[0010] Determine the information loss amount of each local region according to the number of downsampling iterations of each local region; screen out high-printing-information regions from the local regions according to the magnitude of the information loss amount.

[0011] Further, the step of determining the information loss amount of each local region according to the number of downsampling iterations of each local region includes:

[0012] Taking any local region as the target local region, calculate the average value of the difference in the number of downsampling iterations between the target local region and other local regions as the information loss amount of the target local region.

[0013] Further, the step of screening out high-printing-information regions from the local regions according to the magnitude of the information loss amount includes:

[0014] Taking the local regions with information loss amounts greater than the preset loss threshold as high-printing-information regions.

[0015] Further, the step of screening out printing defect regions from the high-printing-information regions according to the edge lines and regional color differences in the high-printing-information regions includes:

[0016] Determine the possibility of edge misprinting according to the repetition of edge lines in the high-printing-information regions.

[0017] Determine the possibility of color difference defect according to the differences in regional color differences in the high-printing-information regions.

[0018] Combine the possibility of edge misprinting and the possibility of color difference defect to obtain the printing defect probability of the high-printing-information regions.

[0019] When the printing defect probability exceeds the preset normal defect range, take the high-printing-information region corresponding to the printing defect probability as the printing defect region.

[0020] Further, the step of determining the possibility of edge misprinting according to the repetition of edge lines in the high-printing-information regions includes:

[0021] Take any edge line in the high-printing information area as the target edge line, and take any other edge line except the target edge line as the edge line to be analyzed;

[0022] Perform linear fitting on the target edge line and the edge line to be analyzed respectively to obtain a target edge fitting line and an edge line to be analyzed fitting line;

[0023] Determine the repetition probability of the corresponding target edge line and the edge line to be analyzed according to the inclination degree and length of the target edge fitting line and the edge line to be analyzed fitting line;

[0024] Combine the repetition probability of the target edge line and other edge lines except the target edge line, and the coefficient of variation on both sides of the target edge line and other edge lines except the target edge line to determine the possibility of line misprint of the target edge line;

[0025] Determine the edge misprint possibility of the high-printing information area according to the misprint possibility of all lines in the high-printing information area.

[0026] Further, the determining the color difference defect possibility according to the difference situation between the regional color differences in the high-printing information area includes:

[0027] Cluster the edge lines according to the lengths of all edge lines in the high-printing information area and the slopes of the corresponding edge fitting lines to obtain multiple clusters;

[0028] For the target edge line, combine the repetition probability between the target edge line and the edge lines within the cluster to which it belongs, and the coefficient of variation on both sides of the target edge line and the edge lines within the cluster to which it belongs to determine the line defect possibility of the target edge line;

[0029] Determine the color difference defect possibility of the high-printing information area according to the defect possibility of all lines in the high-printing information area.

[0030] Further, the comparing the printed defect area with the printing template and identifying the printing defects in the printed defect area includes:

[0031] Use the projection method to project and map the printed defect area onto the printing template image to determine the position of the printed defect area with printing defects on the printing template image;

[0032] Identify the types of printing defects in the printed defect area.

[0033] Further, the locally segmenting the printed product image includes:

[0034] Obtain the peak and valley points in the grayscale histogram of the printed product image;

[0035] Using the pixel points corresponding to the peak-valley points as seed points, multiple local regions are obtained by using the region growing algorithm.

[0036] Further, the multi-round downsampling iteration for different local regions includes:

[0037] Using the image pyramid algorithm to perform downsampling processing on the local regions.

[0038] In a second aspect, a method for fine observation of a printed still image with a real-time self-checking function is provided. The method includes:

[0039] Obtaining an image of a printed product during the printing process;

[0040] Performing local segmentation on the image of the printed product; performing multi-round downsampling iteration on different local regions, analyzing the local information loss of each local region during each downsampling iteration, screening out high-printing information regions; according to the edge lines and regional color differences in the high-printing information regions, screening out printed defect regions from the high-printing information regions;

[0041] Comparing the printed defect regions with a printing template to identify the printed defects in the printed defect regions.

[0042] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor. An executable code is stored in the memory. When the processor executes the executable code, the embodiments of the first aspect are implemented as possible.

[0043] In a fourth aspect, an embodiment of the present invention provides a computer program product, which includes: computer program code. When the computer program code runs on a computer, the computer is caused to execute the method in the first aspect or any possible implementation manner of the first aspect.

[0044] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the embodiments of the first aspect as possible.

[0045] The embodiments of the present invention have at least the following beneficial effects:

[0046] Embodiments of the present invention relate to the field of image processing technology. First, the present invention performs local segmentation on the printed product image. The reason for local segmentation of the image is that not every area in the printed product image will have printing defects, but printing defects may occur in every area. Therefore, in order to achieve faster defect localization, embodiments of the present invention first perform block processing on the entire printed product image. Then, the printing defect area is screened out through the defect screening module. Because the more monotonous the information inside the local area is, the less likely it is that the local area carries printing information, or it only contains a small amount of printing information. When printing defects occur, they are most likely in areas with a large number of printing defects. Therefore, through information loss, areas with a large amount of information can be obtained, and high printing information areas with a high content of printing information can be obtained, reducing the computational complexity when obtaining block printing defects later; secondly, because the high printing information area contains relatively more printing patterns, the possibility of printing defects appearing in it is much greater than that in non-high printing information areas. Common printing defects are often misprints and color differences. Therefore, the defect screening module screens out the printing defect area based on the edge lines and regional color differences in the high printing information area. Identifying printing defects in the printing defect area improves the accuracy of defect identification compared to directly detecting defects in the entire image. Moreover, after determining the printing defect area according to the defect screening module, when only identifying the printing defects in the printing defect area, the computational complexity in the defect identification process is reduced, and the accuracy of defect identification is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 It is a system block diagram of a fine observation system for printed still images with a real-time self-check function provided by an embodiment of the present invention;

[0049] Figure 2 It is a method flow chart of a fine observation method for printed still images with a real-time self-check function provided by an embodiment of the present invention;

[0050] Figure 3 It is a step flow chart implemented by the defect screening module 20 provided by an embodiment of the present invention;

[0051] Figure 4 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0052] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of a fine observation system for printed still images with a real-time self-check function proposed according to the present invention.

[0053] In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0054] Among them, in the description of the embodiments of the present invention, unless otherwise specified, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present invention, "a plurality" means two or more than two.

[0055] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0057] The following describes the embodiments of the present invention in conjunction with the accompanying drawings. Those of ordinary skill in the art will know that with the development of technology and the emergence of new scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0058] The embodiments of the present invention provide a specific implementation method for a fine observation system for printed still images with a real-time self-check function. This method is applicable to the scenario of printed defect detection. In this scenario, the printing equipment has a short period of stillness, that is, intermittent pauses during the operation of the machine. An image acquisition device is used to acquire images of the still printed products during the intermittent pauses.

[0059] The following specifically describes the specific solution of a fine observation system for printed still images with a real-time self-check function provided by the present invention in conjunction with the accompanying drawings.

[0060] Please refer to Figure 1, which shows a system block diagram of a fine observation system for printed still images with a real-time self-check function provided by an embodiment of the present invention. The system includes the following modules:

[0061] An image acquisition module 10, configured to acquire images of printed products during the printing process.

[0062] In an embodiment of the present invention, the self-check of the printing equipment is realized by using the already printed pattern information during the printing process. Before the self-check, it is necessary to collect information on the printing effect. The specific printing effect collection method is as follows:

[0063] First, ensure that the printing equipment has a short-term pause, that is, an intermittent pause during the operation of the machine; then use the image acquisition device to acquire images of the stationary printed products during the intermittent pause to obtain stationary printed images.

[0064] Then, use an image segmentation algorithm to separate the background information from the acquired stationary printed images and only retain the printed patterns.

[0065] Finally, use a grayscale algorithm to grayscale the retained printed patterns to obtain images of printed products.

[0066] Through image segmentation and grayscaling, the acquisition and preprocessing of the images of printed products to be self-checked are completed.

[0067] A defect screening module 20, configured to perform local segmentation on the images of printed products, perform multiple rounds of downsampling iteration on different local regions; analyze the local information loss of each local region during each downsampling iteration, screen out high-printing information regions; and screen out printing defect regions from the high-printing information regions according to the edge lines and regional color differences in the high-printing information regions.

[0068] During the printing process, if a stop self-check is performed for the defect test of printed products, it will obviously delay the production efficiency. Without a stop self-check for fine observation, a large amount of computing power is often required, and equipping higher processing equipment will inevitably increase the cost. Therefore, on the basis of non-stop self-check, the present invention proposes a local analysis method to determine the printing defect regions of printed products, and then uses local regions for fine comparison to complete the self-check of printed products.

[0069] In some embodiments, a high-printing information region is determined from the images of printed products, and a printing defect region is screened out from the high-printing information region. That is, the above-mentioned defect screening module 20 can be realized through Figure 3 the steps shown:

[0070] Step S210: Locally segment the printed product image and perform multiple rounds of downsampling iteration on different local regions.

[0071] An image acquisition device is used to obtain the printed product image during the printing process. In the printed product image, not every region will have printing defects, but printing defects may occur in every region. Therefore, in order to achieve faster defect localization, in the embodiments of the present invention, the overall printed product image is first divided into blocks. Specifically:

[0072] Obtain the grayscale histogram of the printed product image.

[0073] Use the method of local extrema to obtain the peak-valley points in the grayscale histogram of the printed product image; form a peak-valley point sequence , and each grayscale value in the peak-valley point sequence belongs to .

[0074] Use each pixel point corresponding to a peak-valley point as a seed point on the printed product image and perform processing using the region growing algorithm to achieve local segmentation of the printed product image, obtaining A blocks, and taking the blocks as local regions.

[0075] The logic of selecting peak-valley points for region growing here is that the peak-valley points are the basic background of most printed products and a small number of pixel points with grayscale values in the non-background regions. Segmenting the printed product image with these types of pixel points can better classify most of the background regions into the same unit, and for a small number of pixel points with grayscale values in the non-background regions, similarly, the non-background regions can be effectively divided.

[0076] Step S220: Perform multiple rounds of downsampling iteration on different local regions, analyze the local information loss of each local region during each downsampling iteration, and screen out high-printing-information regions.

[0077] Then, estimate the information loss amount for each block. The basic logic is as follows: The more monotonic the information within a local region is, the less likely it is that the local region contains printing information, or it only contains a small amount of printing information. When printing defects occur, they are most likely in regions with a large number of printing defects. Therefore, through information loss, regions with a large amount of information can be obtained, reducing the computational amount when obtaining block printing defects later.

[0078] First, perform multiple rounds of downsampling iteration on different local regions. In the embodiments of the present invention, the image pyramid algorithm is used to perform downsampling processing on the local regions.

[0079] Furthermore, analyze the local information loss of each local region during each downsampling iteration and screen out high-printing-information regions:

[0080] Calculate the structural similarity of the regions before and after downsampling for each local region, perform a negative correlation mapping process on the structural similarity to obtain the local information loss of each local region; for each local region, when the local information loss of the local region exceeds the preset normal loss range, stop the downsampling iteration; otherwise, when the local information loss of the local region belongs to the preset normal loss range, continue to perform downsampling on the local region. In the embodiment of the present invention, the preset normal loss range is , and in other embodiments, it can also be adjusted by the implementer according to the actual situation. Among them, the negative correlation mapping of the structural similarity can be achieved by subtracting the structural similarity from the constant 1.

[0081] After performing downsampling on each local region, determine the information loss amount of each local region according to the number of downsampling iterations of each local region. The larger the information loss amount, it indicates that the th local region has more complex printed information and more information compared to the printed information of the remaining blocks, and at the same time, it also indicates that the th local region has a greater possibility of printing defects.

[0082] Take any local region as the target local region, and calculate the average value of the difference in the number of downsampling iterations between the target local region and other local regions as the information loss amount of the target local region.

[0083] In some embodiments, taking the a-th local region as the target local region, the information loss amount of the target local region is calculated by the formula: ; where is the number of downsampling times of the a-th local region; is the number of downsampling times of the a'-th local region except for the a-th local region; A is the number of other local regions except for the a-th local region. ; , and there is always .

[0084] Regard the local regions with information loss amount greater than the preset loss threshold as high printed information regions. Because the larger the information loss amount, the more complex the corresponding printed information is and the greater the probability of printing defects.

[0085] In the embodiment of the present invention, the method for determining the preset loss threshold is: taking the median of the information loss amounts corresponding to the local regions as the preset loss threshold.

[0086] In some embodiments of the present invention, high-printing-information blocks can also be obtained by printing information content sequences. The specific obtaining method is as follows: all local regions are tagged with information loss amounts, and sorted in descending order through the tags to obtain a printing information content sequence. The higher the ranking in the printing information content sequence, the more complex the printing information contained, and the greater the probability of printing defects. Select the first 50% of the local regions in the printing information content sequence as high-printing-information regions.

[0087] Step S230, according to the edge lines and regional color differences in the high-printing-information regions, screen out the printing defect regions from the high-printing-information regions.

[0088] In steps S210 to S220, the downsampling algorithm is used to determine the information loss of local regions in the printed product image, and high-printing-information regions with relatively high printing information content are obtained. Because the high-printing-information regions contain relatively more printing patterns, the possibility of printing defects appears much greater than that of non-high-printing-information regions. Common defects in paper printing are often misprints and color differences. Therefore, the present invention analyzes the edges and gray values in the high-printing-information regions to obtain the probability of defects in each high-printing-information region.

[0089] First, according to the repetition of edge lines in the high-printing-information regions, determine the possibility of edge misprinting. The specific obtaining method is as follows:

[0090] Taking the b-th high-printing-information region as an example, the printing defect probability of the specific defect probability The obtaining method is as follows:

[0091] First, for all edges of the b-th high-printing-information region, obtain the possibility of edge misprinting. The specific obtaining logic is as follows:

[0092] Misprinting refers to the phenomenon that when the printing equipment prints a pattern at a specific position, due to the movement of the paper or the mutual friction between the equipment and the paper, misaligned printing or ghosting occurs. The most obvious feature of misprinting is the appearance of repeated edges, and because the patterns near the repeated edges are the same, the colors expressed are the same. Therefore, based on this feature, the possibility of edge misprinting can be obtained. The specific obtaining method is as follows:

[0093] First, use the edge detection algorithm to perform edge detection on the b-th high-printing-information region to obtain all edge lines of the b-th high-printing-information region.

[0094] Then, obtain the repetition probability of each edge line. Take any edge line in the high-printing information area as the target edge line, and take any other edge line except the target edge line as the edge line to be analyzed. In the embodiment of the present invention, take the th edge line among all the edge lines in the bth high-printing information area as the target edge line, and take the th edge line as the edge line to be analyzed as an example. The method for obtaining the corresponding repetition probability is as follows:

[0095] Perform linear fitting on the target edge line and the edge line to be analyzed respectively to obtain the target edge fitting line and the edge line to be analyzed fitting line; in the embodiment of the present invention, for the th edge line and the th edge line, perform linear fitting respectively to obtain the target edge fitting line after fitting the th edge line and the edge line to be analyzed fitting line .

[0096] Because there are various forms of the th edge line and the th edge line, such as the irregular bending of the printed pattern, it is not convenient to calculate. Therefore, in the embodiment of the present invention, by performing linear fitting on the edge line, the difference of the straight line is analyzed to map the similarity of the edge. Because misprint is the same printed pattern printed multiple times, the more the slopes of the fitting lines and the lengths of the edge lines are the same, the greater the possibility that the edge lines corresponding to these two fitting lines are misprinted. Therefore, determine the repetition probability of the corresponding target edge line and the edge line to be analyzed according to the inclination degree and length of the target edge fitting line and the edge line to be analyzed fitting line.

[0097] First, perform difference calculation on the target edge fitting line and the edge line to be analyzed fitting line : ; where is the slope of the target edge fitting line corresponding to the th edge line, is the slope of the edge line to be analyzed fitting line corresponding to the th edge line, is the length of the th edge line, the The length of an edge line, where the length of the edge line is the number of pixels of the edge line; in the embodiments of the present invention, the slope of the edge fitting line is used to characterize the inclination degree of the edge fitting line; and The greater the difference, the smaller the similarity between the th edge line and the th edge line, and vice versa. Therefore, the similarity between the th edge line and the th edge line can be obtained as the repetition probability .

[0098] Using the above method, the repetition probability of the target edge line and all the remaining edge lines to be analyzed except the target edge line can be obtained.

[0099] Finally, combining the repetition probability of the target edge line and other edge lines except the target edge line, and the coefficient of variation on both sides of the target edge line and other edge lines except the target edge line, determine the possibility of line misprinting of the target edge line. Specifically:

[0100] Using the repetition probability of the th edge line and all the remaining th edge lines, obtain the possibility of line misprinting of the th edge line , which is specifically as follows:

[0101] ;

[0102] Among them, and respectively represent the coefficients of variation of the gray values of the pixel points on both sides of the th edge line and the th edge line.

[0103] Among them, the method for obtaining the coefficient of variation of the gray values of the pixel points on both sides of the edge line is: taking the left side of the edge line as the target side, taking the coefficient of variation of the gray values of the pixel points adjacent to the pixel points in the edge line in the target side as the left coefficient of variation, and averaging the left and right coefficients of variation of the edge line to obtain the coefficient of variation of the edge line.

[0104] The ratio of the coefficients of variation of two edge lines is used to measure the logic of the similarity degree of the pixel point gray values on both sides of the th edge line and the th edge line; the th edge line and the The more similar the grayscale values of the pixel points on both sides of an edge line are, the corresponding is smaller, and vice versa, the corresponding is larger. Because misprinting means the same printing information is printed incorrectly multiple times, then when there are more edge lines similar to the th edge line among all the edge lines, and the grayscale representation forms on both sides of the edge lines are more similar, it indicates that the printing information corresponding to the th edge line is more likely to be misprinted, and vice versa.

[0105] Using the above method, the line misprinting possibility of each edge line in the B high-printing-information regions can be obtained.

[0106] According to the line misprinting possibilities of all the lines in the high-printing-information region, determine the edge misprinting possibility of the high-printing-information region. Specifically: for any high-printing-information region ui, calculate the average value of the line misprinting possibilities of all the lines in the high-printing-information region ui as the edge misprinting possibility of the high-printing-information region ui.

[0107] Second, according to the difference situation between the regional color differences in the high-printing-information region, determine the color difference defect possibility.

[0108] Color difference occurs during printing because the ink is uneven or the force applied by the equipment is not unified enough, resulting in different information expressions of the same pattern in different regions. Therefore, based on this logic, the embodiments of the present invention determine the color difference defect possibility of the high-printing-information region.

[0109] Therefore, according to the lengths of all the edge lines in the high-printing-information region and the slopes of the corresponding edge fitting straight lines, cluster the edge lines to obtain multiple clusters; for the target edge line, combine the repetition probability between the target edge line and the edge lines in the cluster it belongs to, and the coefficient of variation on both sides of the target edge line and the edge lines in the cluster it belongs to, to determine the line defect possibility of the target edge line; according to the line defect possibilities of all the lines in the high-printing-information region, determine the color difference defect possibility of the high-printing-information region. Specifically: take the average value of all the line defect possibilities in the high-printing-information region to obtain the color difference defect possibility of the high-printing-information region.

[0110] Obtain the color difference defect possibility of each high-printing-information region. Specifically, taking the bth high-printing-information region as an example, the method for obtaining the color difference defect possibility of its existence of color difference defects is as follows:

[0111] First, for each high-printing-information region, obtain the edges to get multiple edge lines.

[0112] Next, for each high-printing information area, the edge fitting straight line corresponding to the edge line in each high-printing information area is obtained by using the straight line fitting method.

[0113] The slope of the edge fitting straight line corresponding to the edge line in all high-printing information areas and the length of the edge line are used as the two-dimensional parameters of each edge line;

[0114] Then, taking the b-th high-printing information area as an example, the K clustering of the two-dimensional parameters of slope-length is performed on each edge line in the b-th high-printing information area by using the two-dimensional K clustering algorithm, and the cluster corresponding to each edge in the b-th high-printing information area is obtained;

[0115] Next, for each edge line in the b-th high-printing information area, the possibility of color difference defects is obtained according to the cluster corresponding to the edge. Here, still taking the -th edge line in the b-th high-printing information area as the target edge line, the possibility of color difference defects of the target edge line is calculated by the formula:

[0116] ;

[0117] Among them, represents the repetition probability of the -th edge line in the b-th high-printing information area and the -th edge line in the cluster to which the -th edge line belongs; is the coefficient of variation of the gray values of the pixel points on both sides of the -th edge line in the b-th high-printing information area; is the coefficient of variation of the gray values of the pixel points on both sides of the -th edge line in the cluster to which the -th edge line belongs; is the number of edge lines in the cluster to which the -th edge line belongs; is the number of edge lines in the b-th high-printing information area; is the line defect possibility of the -th edge line in the b-th high-printing information area.

[0118] The greater the difference in the performance of the gray values of the pixel points on both sides of the edge lines in the b-th high-printing information area and the edge lines in the same cluster in the remaining high-printing information areas, the greater the possibility of color difference, and vice versa.

[0119] The greater the probability that the b-th high-printing information area has misprinting or color difference defects, the greater the probability that the b-th high-printing information area has printing defects, and vice versa.

[0120] Therefore, finally, by combining the edge misprinting possibility and the color difference defect possibility, the printing defect probability of the high-printing information area is obtained. Both the edge misprinting possibility and the color difference defect possibility are positively correlated with the printing defect probability. In an embodiment of the present invention, the normalized value of the sum of the edge misprinting possibility and the color difference defect possibility is used as the printing defect probability of the high-printing information area.

[0121] Printing defect probability The calculation formula is:

[0122] ; where Norm is the normalization function.

[0123] By using the above method, the printing defect probability of each high-printing information area having printing defects can be obtained, and then based on the printing defect probability, the printing defect areas are screened out from the high-printing information areas.

[0124] As an embodiment of the present invention, when the printing defect probability exceeds the preset normal defect range, the high-printing information area corresponding to the printing defect probability is used as the printing defect area.

[0125] Specifically: for all high-printing information areas, a descending order sequence of printing defects can be obtained by sorting in descending order according to the size of the printing defect probability of having printing defects; then, using the threshold screening method, the areas with printing defects in the descending order sequence of printing defects are obtained by using the custom parameter method; specifically, the custom parameter method is to give a preset normal defect range through experience. When the value in the descending order sequence of printing defects is greater than the preset normal defect range, it is considered as a high-printing information block with defects; otherwise, it is the opposite; the preset normal defect range adopted in the present invention is [0 - 0.9], and this preset normal defect range can be adjusted according to the actual situation. The larger the parameter value of the right endpoint of the preset normal defect range, the more accurate the screening, but the corresponding calculation amount is larger.

[0126] Through the preset normal defect range, the printing defect areas existing in the printing process are screened out.

[0127] The defect determination module 30 is used to compare the printing defect area with the printing template and identify the printing defects in the printing defect area.

[0128] By comparing the printing defect area with a probability of having defects in the printed product image with the printing template, the printing defects are located and the electrical appliance is controlled to complete the printing production.

[0129] The printed defect areas where printing defects exist during the printing process are determined by the image acquisition module 10 and the defect screening module 20 above. Now, these printed defect areas are compared with the corresponding printing templates to locate the printing defects and control the printing equipment to complete the printing production. The specific method is as follows:

[0130] 1. First, use the projection method to project and map all printed defect areas onto the printing template image to determine the positions of the printed defect areas with printing defects on the printing template image;

[0131] 2. Then, use the existing defect recognition model to identify the types of printing defects and retrieve the causes of defects in the printed defect areas on the printing template image;

[0132] 3. Finally, adjust the corresponding areas and the printing equipment according to the types of defects and the causes of defects to complete the printing production.

[0133] Please refer to Figure 2 , which shows a method flow chart of a method for fine observation of a printed still image with a real-time self-check function provided by an embodiment of the present invention. The method includes the following steps:

[0134] Step S100, obtaining a printed product image during the printing process;

[0135] Step S200, performing local segmentation on the printed product image; performing multiple rounds of downsampling iteration on different local areas, analyzing the local information loss of each local area during each downsampling iteration, screening out high-printing information areas; and screening out printed defect areas from the high-printing information areas according to the edge lines and regional color differences in the high-printing information areas;

[0136] Step S300, comparing the printed defect areas with the printing template to identify the printing defects in the printed defect areas.

[0137] Optionally, the transmission medium may be a wired link, such as but not limited to, coaxial cable, optical fiber, and digital subscriber line, etc., or a wireless link, such as but not limited to, Wireless Fidelity (WIFI), Bluetooth, and mobile device network, etc.

[0138] It should be noted that: for the device provided in the above embodiment, only the above-mentioned division of each functional module is used for illustration. In actual application, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above.

[0139] Figure 4 This is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Exemplarily, as Figure 4 shown, the computer device 400 includes: a memory 410, a processor 420, and a computer program 430 stored in the memory 410 and running on the processor 420. When the processor 420 executes the computer program 430, the computer device can execute any of the aforementioned printing still image fine observation systems with real-time self-checking functions.

[0140] In addition, an embodiment of the present invention also protects a device, which may include a memory and a processor. Among them, an executable program code is stored in the memory, and the processor is used to call and execute the executable program code to execute a printing still image fine observation system with real-time self-checking functions provided by an embodiment of the present invention.

[0141] An embodiment of the present invention can divide the functions of the device according to the above method examples. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0142] In the case of dividing each module according to each function, the device may further include a signal uploading module, a determination module, an adjustment module, etc. It should be noted that all relevant contents of each step involved in the above method embodiment can be cited in the function description of the corresponding functional module, and will not be repeated here.

[0143] It should be understood that the device provided by an embodiment of the present invention is used to execute the above printing still image fine observation system with real-time self-checking functions, so the same effect as the above implementation method can be achieved.

[0144] In the case of adopting an integrated unit, the device may include a processing module and a storage module. Among them, when the device is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device to execute mutual program codes, etc. Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logical blocks, modules, and circuits described in combination with the disclosure of the present invention. The processor can also be a combination of computing functions, such as including a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module can be a memory.

[0145] In addition, the device provided by the embodiments of the present invention may specifically be a chip, a component or a module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip may execute a fine observation system for printed still images with a real-time self-checking function provided by the above embodiments.

[0146] The embodiments of the present invention also provide a computer-readable storage medium. Computer program code is stored in the computer-readable storage medium. When the computer program code runs on a computer, the computer is enabled to execute the above related method steps to implement a fine observation system for printed still images with a real-time self-checking function provided by the above embodiments.

[0147] The embodiments of the present invention also provide a computer program product. When the computer program product runs on a computer, the computer is enabled to execute the above related steps to implement a fine observation system for printed still images with a real-time self-checking function provided by the above embodiments.

[0148] Among them, the device, the computer-readable storage medium, the computer program product or the chip provided by the embodiments of the present invention are all used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be elaborated here. Through the description of the above embodiments, those skilled in the art can understand that for the convenience and conciseness of description, only the above division of each functional module is used as an example. In actual applications, the above functions may be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by the present invention, it should be understood that the disclosed device and method may be implemented in other ways.

[0149] The device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other may be through some interfaces. The indirect coupling or communication connection of the device or unit may be in an electrical, mechanical or other form.

[0150] It should also be noted that in this text, the terms "comprising", "including" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or terminal device comprising a series of elements not only includes those elements but also other elements not explicitly listed, or further includes elements inherent to such a process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising said element.

[0151] It should be noted that the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0152] Each embodiment in this specification is described in a progressive manner, and the same or similar parts among the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0153] The above content is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A fine observation system for printed still images with a real-time self-check function, characterized in that, The system includes the following modules: An image acquisition module, configured to obtain an image of a printed product during the printing process; A defect screening module, configured to perform local segmentation on the image of the printed product; Perform multiple rounds of downsampling iterations on different local regions, analyze the local information loss of each local region during each downsampling iteration, and screen out high-printing-information regions; according to the edge lines and regional color differences in the high-printing-information regions, screen out printing defect regions from the high-printing-information regions; wherein, the method for obtaining the high-printing-information regions is: calculate the structural similarity of the regions before and after downsampling of each local region, perform a negative correlation mapping process on the structural similarity to obtain the local information loss of each local region; for each local region, when the local information loss of the local region exceeds the preset normal loss range, stop the downsampling iteration; determine the information loss amount of each local region according to the number of downsampling iterations of each local region; screen out high-printing-information regions from the local regions according to the magnitude of the information loss amount; A defect determination module, configured to compare the printing defect regions with a printing template to identify printing defects in the printing defect regions.

2. The fine observation system for printed still images with a real-time self-check function according to claim 1, characterized in that, The determining the information loss amount of each local region according to the number of downsampling iterations of each local region includes: Taking any local region as the target local region, calculating the average value of the difference in the number of downsampling iterations between the target local region and other local regions as the information loss amount of the target local region.

3. The fine observation system for printed still images with a real-time self-check function according to claim 1, characterized in that, The screening out high-printing-information regions from the local regions according to the magnitude of the information loss amount includes: Taking the local regions with information loss amounts greater than the preset loss threshold as high-printing-information regions.

4. The fine observation system for printed still images with a real-time self-check function according to claim 1, characterized in that, The screening out printing defect regions from the high-printing-information regions according to the edge lines and regional color differences in the high-printing-information regions includes: Determining the possibility of edge misprinting according to the repetition situation between the edge lines in the high-printing-information regions; Determining the possibility of color difference defects according to the difference situation between the regional color differences in the high-printing-information regions; Combining the possibility of edge misprinting and the possibility of color difference defects to obtain the printing defect probability of the high-printing-information regions; When the printing defect probability exceeds the preset normal defect range, taking the high-printing-information region corresponding to the printing defect probability as the printing defect region.

5. The fine observation system for printed still images with real-time self-checking function according to claim 4, characterized in that, The determining the possibility of edge misprinting according to the repetition situation between the edge lines in the high-printing-information regions includes: Taking any edge line in the high-printing-information region as the target edge line, and taking any other edge line except the target edge line as the edge line to be analyzed; Performing linear fitting on the target edge line and the edge line to be analyzed respectively to obtain a target edge fitting line and an edge line to be analyzed fitting line; Determining the repetition probability of the corresponding target edge line and the edge line to be analyzed according to the inclination degree and length of the target edge fitting line and the edge line to be analyzed fitting line; Determine the likelihood of misprint of the target edge line by combining the repetition probability of the target edge line and other edge lines except the target edge line, and the coefficient of variation on both sides of the target edge line and other edge lines except the target edge line; Determine the likelihood of edge misprint in the high-printing-information area based on the likelihood of misprint of all lines in the high-printing-information area.

6. The fine observation system for printed still images with a real-time self-check function according to claim 5, characterized in that, The determining the likelihood of chromatic aberration defect according to the difference situation between the regional chromatic aberrations in the high-printing-information area includes: Cluster the edge lines according to the lengths of all edge lines in the high-printing-information area and the slopes of the corresponding edge fitting lines to obtain multiple clusters; For the target edge line, determine the likelihood of line defect of the target edge line by combining the repetition probability between the target edge line and the edge lines within the cluster to which it belongs, and the coefficient of variation on both sides of the target edge line and the edge lines within the cluster to which it belongs; Determine the likelihood of chromatic aberration defect in the high-printing-information area based on the likelihood of line defect of all lines in the high-printing-information area.

7. The fine observation system for printed still images with a real-time self-check function according to claim 1, characterized in that, The comparing the printed defect area with the printing template and identifying the printing defects in the printed defect area includes: Use the projection method to project and map the printed defect area onto the printing template image to determine the position of the printed defect area with printing defects on the printing template image; Identify the types of printing defects in the printed defect area.

8. The fine observation system for printed still images with a real-time self-check function according to claim 1, characterized in that, The locally segmenting the printed product image includes: Obtain the peak-valley points in the grayscale histogram of the printed product image; Use the region growing algorithm with the pixel points corresponding to the peak-valley points as seed points to obtain multiple local areas.

9. The fine observation system for printed still images with a real-time self-check function according to claim 1, characterized in that, The performing multiple rounds of downsampling iteration on different local areas includes: Use the image pyramid algorithm to perform downsampling processing on the local areas.

Images