CTP plate making quality detection and analysis method based on image recognition

Abstract

The invention discloses a CTP plate-making quality detection and analysis method based on image recognition, and relates to the field of CTP plate-making quality detection based on image recognition, and the method comprises the following steps: synchronously collecting a manufactured CTP printing plate image and a printed matter image produced by the printing plate by using a high-resolution image collection device; constructing a unified coordinate system by taking the geometric center and the angular points outside the effective printing area of the printing plate as physical reference marks, and performing feature positioning and area division based on the unified coordinate system; the deviation data set is generated by extracting multi-dimensional feature data such as the dot expansion rate and the dirty point coverage rate, combinatorial logic analysis is carried out in combination with the feature type and the position relation to judge the reason of dot excessive expansion and the source of dirty point generation, automatic detection and source tracing of the CTP plate making quality are achieved, and the detection efficiency is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of CTP platemaking quality based on image recognition, and relates to the technical field of a CTP platemaking quality detection and analysis method based on image recognition. Background Art

[0002] CTP, short for Computer to Plate, is a modern platemaking process that outputs digital graphics directly onto printing plates via lasers or other imaging methods. It is currently widely used in the printing industry. During the CTP platemaking process, plate quality indicators such as dot accuracy and contamination directly determine the clarity, color consistency, and cleanliness of the final print. Defects in the plate can lead to the scrapping of batches of printed products, resulting in significant economic losses. Image recognition-based quality monitoring and analysis has become a core technology in the CTP platemaking process.

[0003] At present, existing technologies have proposed CTP platemaking quality detection schemes for image recognition. For example, the invention patent with publication number CN101556251A proposes a CTP platemaking quality detection method based on a digital signal processor. It uses a specific structure to capture images of dots, lines, and chessboard areas on the CTP printing plate measurement and control strip, and obtains microscopic imaging at a fixed object distance by adjusting relevant components. The central sub-image of the captured image is intercepted, and then the dot features of the ladder area, the straight line features of the straight line area, and the chessboard block features of the chessboard area are extracted respectively. A comprehensive evaluation is performed based on the membership function of each feature combined with the weight, and the results are displayed on an LCD screen, realizing objective detection of CTP platemaking quality, replacing traditional manual subjective evaluation, and improving detection efficiency and consistency.

[0004] Although the above-mentioned existing solutions have achieved certain results in CTP platemaking quality inspection based on image recognition, they still have the following shortcomings: First, the image on the printed product comes entirely from the printing plate. Any abnormal point on the printed product is caused by the printing plate or the printing process. The existing technology only collects and detects the image of the printing plate itself, which is an isolated quality control link. As a result, it is impossible to trace the quality through the mapping relationship between the printing plate and the printed product. It is impossible to accurately determine whether the quality problems in the printing process are caused by the platemaking link or the printing link, which reduces the comprehensiveness and accuracy of the quality analysis.

[0005] Secondly, the existing technology only obtains a fixed object distance for image acquisition by adjusting the object distance fine-tuning ring. The statistical measurement is based on a small piece of the intercepted sub-image. The feature analysis of each area is independent of each other and lacks position correlation. As a result, after the feature abnormality is detected, the specific position of the abnormality on the printing plate cannot be accurately located, which brings great difficulties to the targeted repair of the printing plate and the elimination of unqualified areas, reducing the utilization rate of the printing plate.

[0006] In addition, the existing technology can only give a binary judgment of qualified or unqualified by comparing the extracted dot, line, and chessboard block features with the membership function, and generate a defect report pointing to a specific process link. After detecting quality abnormalities, it is necessary to check multiple possible influencing factors such as exposure parameters, plate quality, equipment accuracy, and operating procedures one by one. It is impossible to quickly identify the root cause of the problem, thereby reducing overall production efficiency. Summary of the Invention

[0007] In view of this, in order to solve the problems raised in the above background technology, the present invention provides a technical field of a CTP platemaking quality detection and analysis method based on image recognition.

[0008] The purpose of the present invention can be achieved through the following technical solution: a CTP platemaking quality detection and analysis method based on image recognition, comprising: S1, using a high-resolution image acquisition device to synchronously capture the image of a manufactured CTP printing plate and the image of a printed product produced by the printing plate.

[0009] S2. After pre-processing the collected printing plate and printed matter images, a unified coordinate system of the printing plate image and the printed matter image is established based on the physical correspondence between the printing plate and the printed matter.

[0010] S3. Identify and segment the single dot area and the dirty dot area on the two images respectively, and extract the geometric features of the dot and the shape features of the dirty dot to form a printing plate feature data set and a printed product feature data set.

[0011] S4. Based on a unified coordinate system, the coordinates of the geometric features of the dots in the printed feature dataset are mapped and matched with the printing plate feature dataset, the dot enlargement rate of each corresponding dot is calculated one by one, and the dirty dot coverage rate of the corresponding area is counted to generate a deviation dataset containing position information.

[0012] S5. Generate a quality defect root cause quality report based on the deviation data set combined with the combination logic of feature position relationship and feature type.

[0013] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention uses high-resolution image acquisition equipment to synchronously capture the image of the manufactured CTP printing plate and the image of the printed product produced by the printing plate. It can trace the quality of the printing plate and the printed product through correlation analysis, and accurately judge whether the quality problem originates from the platemaking link or the printing link, effectively solving the problem of insufficient comprehensiveness and accuracy of quality analysis.

[0014] (2) The present invention uses the geometric center and corner points outside the effective printing area of ​​the printing plate as physical reference marks to construct a unified coordinate system, and performs feature positioning and area division based on the unified coordinate system, so that the feature analysis of each area has a clear position correlation, which greatly facilitates the targeted repair of the printing plate and the elimination of unqualified areas, and improves the utilization rate of the printing plate and the detection efficiency.

[0015] (3) The present invention generates a deviation data set by extracting multi-dimensional feature data such as dot expansion rate and dirty dot coverage rate, and performs combined logic analysis based on the feature type and position relationship to determine the cause of excessive dot expansion and the root cause of dirty dots. There is no need to check multiple influencing factors one by one, and the root cause of quality defects can be quickly identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 This is a diagram of the steps for implementing the method of the present invention.

[0018] Figure 2 A flow chart is provided for establishing a unified coordinate system for the CTP printing plate and printed products of the present invention.

[0019] Figure 3 This is a flow chart of binary image dirty point detection and feature extraction of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1 As shown, the present invention provides a CTP platemaking quality detection and analysis method based on image recognition, comprising: S1, using a high-resolution image acquisition device to synchronously acquire an image of a manufactured CTP printing plate and an image of a printed product produced by the printing plate.

[0022] It should be noted that high-resolution image acquisition equipment can use an industrial area array camera, combined with a white LED flat-panel light source to adjust the light source angle and intensity to ensure that the captured image is free of reflections and shadows.

[0023] A pre-made CTP plate is mounted on a high-precision stage, ensuring it is flat and wrinkle-free. A printed product produced using the same plate under standard printing process parameters is then selected and mounted adjacent to the same stage. A trigger controller controls the camera to synchronously capture high-resolution images of both the plate and the printed product. Synchronous acquisition avoids image distortion caused by time differences or interference from environmental changes, improving comparison accuracy.

[0024] S2. After pre-processing the collected printing plate and printed matter images, a unified coordinate system of the printing plate image and the printed matter image is established based on the physical correspondence between the printing plate and the printed matter.

[0025] It should be noted that image preprocessing includes but is not limited to denoising, grayscale, contrast enhancement, etc., so as to improve image quality, unify processing standards, and provide reliable basic data for subsequent core detection steps.

[0026] See also Figure 2 As shown, a specific implementation process of establishing a unified coordinate system for the printing plate image and the printed product image is as follows: when making the plate, the geometric center and corner points outside the effective printing area of ​​the printing plate are marked as physical reference marks.

[0027] A number of non-collinear physical reference marks are identified in the pre-processed printing plate image and the printed product image respectively.

[0028] It should be noted that the use of several reference marks effectively reduces the impact of single mark recognition errors on the overall conversion accuracy, making the coordinate mapping between the printing plate and the printed product more stable and reliable.

[0029] The pixel coordinates of each physical reference mark in the printing plate image and the corresponding pixel coordinates in the printed product image are obtained.

[0030] Based on the two sets of pixel coordinates of the physical reference marks in the printing plate image and the printed product image, a coordinate transformation matrix from the printed product image coordinates to the printing plate image coordinates is constructed, and the pixel coordinates of all feature points in the printed product image are uniformly transformed into the coordinate system of the printing plate image.

[0031] It should be noted that the specific implementation steps for constructing the coordinate conversion matrix are as follows: recording the pixel coordinates of the reference mark in the printing plate image and the pixel coordinates of the corresponding reference mark in the printed product image.

[0032] The affine transformation model is used to solve the transformation matrix between two sets of coordinates through the least squares method, which includes rotation, translation, and scaling parameters.

[0033] The vertices of the plate image are selected as the origin of the unified coordinate system.

[0034] Specifically, the coordinates of all pixels in the printed image are converted into coordinates in the printing plate image coordinate system by performing a dot multiplication operation with the transformation matrix, and the upper left corner vertex of the printing plate image is used as the origin of the unified coordinate system.

[0035] It should be noted that all feature points in the printed image are converted to the coordinate system of the printing plate image so that subsequent feature matching, deviation analysis and other operations can be performed in the same space, which facilitates the unified management of position information and avoids matching errors caused by inconsistent coordinate systems.

[0036] S3. Identify and segment the single dot area and the dirty dot area on the two images respectively, and extract the geometric features of the dot and the shape features of the dirty dot to form a printing plate feature data set and a printed product feature data set.

[0037] A specific implementation step of segmenting the single dot area and the dirty dot area is as follows: performing connected area detection on the pre-processed printing plate image and printed product image, and calculating the average grayscale value of the pixels in each connected area.

[0038] The specific steps of the connected area detection are as follows: first, all pixels of the preprocessed printing plate image and printed product image are traversed row by row and column by column, and the adjacent pixels of each unmarked pixel are judged. Usually, the 8-neighborhood rule is adopted, that is, whether the pixels in the upper, lower, left, right and four diagonal directions meet the connectivity conditions with the pixel, that is, whether the difference in the grayscale values ​​of adjacent pixels is within a preset range.

[0039] Next, if a pixel and its adjacent pixels meet the connectivity condition, these pixels are classified as the same connected area, and a unique identification number is assigned to the area. The marked pixels are no longer detected repeatedly until all pixels are traversed, completing the preliminary identification of the connected area of ​​the entire image.

[0040] Finally, the initially identified connected areas are checked to see if the same area is split due to the pixel traversal order, and the adjacent small areas that meet the merging conditions are merged.

[0041] If the grayscale mean falls within the preset dot grayscale value interval, the connected area is marked as a dot area and divided into independent dot units.

[0042] It should be noted that the preset dot grayscale value interval is determined by performing statistical analysis on a large number of known samples to determine the concentrated distribution range of the dot grayscale values.

[0043] In a grayscale image, the grayscale value of a pixel represents the brightness or darkness of that point. In printing, halftone dots are areas where ink is deposited, and their reflectivity is low, resulting in darker grayscale in the grayscale image. The blank substrate of the printing plate has a high reflectivity and appears brighter in the image. Transforming the image segmentation problem into a statistical judgment problem based on data avoids subjective assumptions.

[0044] The pre-processed plate image and printed product image are converted into grayscale images and then binarized.

[0045] The area with a grayscale value greater than the preset grayscale threshold after binarization is divided into the foreground pixel area, otherwise it is divided into the background pixel area.

[0046] It's important to note that foreground pixels, also known as dirty pixels, are typically foreign matter such as oil, dust, scratches, and ink. These pixels physically block light reflection or absorb more light, resulting in a darker image with a lower grayscale value. Background pixels, representing normal printing plate substrates or printed paper, have a smooth surface and high reflectivity, resulting in a brighter image with a higher grayscale value. This inherent grayscale difference between dirty pixels and background pixels is exploited to distinguish them.

[0047] The preset grayscale threshold traverses all possible thresholds, calculates the intra-class variance of the foreground and background pixels after segmentation based on the threshold, and selects the threshold that maximizes the inter-class variance as the optimal threshold.

[0048] If the area of ​​the foreground pixel region exceeds the noise threshold, it is marked as a dirty point region, and the dirty point region is segmented to generate independent dirty point units.

[0049] It should be noted that dirty spots usually have a certain physical size, which is reflected in the image as a continuous area composed of more pixels. Their area, that is, the number of pixels, will exceed a minimum threshold that can be regarded as a defect.

[0050] The noise threshold is determined based on the theoretical resolution of the imaging system and the statistical analysis of a large number of good image noises.

[0051] The steps of extracting the dot geometric features are as follows: marking a connected area whose grayscale mean falls within a preset dot grayscale value interval as a target connected area.

[0052] The total number of pixels contained in the target connected domain is counted to determine the area parameter of the dot.

[0053] The geometric center of the connected domain is calculated by the pixel coordinates of the contour boundary, and the position coordinates of the center in the unified coordinate system are recorded.

[0054] Specifically, let the connected domain be a finite two-dimensional discrete point set, and the area of ​​the connected domain is , where R represents the set of pixel coordinates of the connected domain, N represents the cardinality of the set, and i represents the i-th pixel. .

[0055] The sum of the x-coordinate components , where Represents the column coordinate of the i-th pixel, the sum of the y coordinate components , where Represents the row coordinate of the i-th pixel.

[0056] Geometric center horizontal coordinate , the vertical coordinate of the geometric center .

[0057] The area parameters and center coordinates of the dots are defined as the dot geometric features.

[0058] In the printing industry, dot gain is one of the most important indicators for measuring print quality. It is directly reflected in the increase in dot area. Taking area as the primary characteristic directly reflects the core issue of quality inspection.

[0059] The center coordinates provide the absolute position information of each dot, which is the basis for subsequent mapping and matching between the printing plate and the printed image.

[0060] See also Figure 3 As shown, the steps of extracting the shape features of dirty points are as follows: scanning the binary image to identify the continuous area formed by all foreground pixels, and treating each continuous area as a candidate dirty point.

[0061] Contour tracing is performed on each candidate dirty point to obtain its boundary contour pixel set.

[0062] Specifically, we traverse the entire image from left to right and top to bottom, finding the first pixel with a value of 1. This point must be one of the upper left boundary points of a connected region, and use it as the starting point for contour tracing. We set an initial movement direction, usually assuming that the foreground point is entering from the left background, so the initial search direction is upward.

[0063] Starting from the current boundary point and the current movement direction, the system checks the pixels in its 8-neighborhood in a clockwise or counterclockwise order, starting from the direction immediately before the current movement direction. For example, if the previous movement was to the right, this time the system starts from the top right and moves clockwise. Once a foreground pixel is found in the neighborhood, it is considered the next boundary point. The coordinates of this new point are recorded in the contour set, and the current point and the current movement direction are updated.

[0064] Repeat the above steps until the following two conditions are met at the same time, then stop tracking: the next boundary point returns to the starting point, and the next movement direction is the same as the initial movement direction.

[0065] At this point, all the recorded point coordinates constitute an ordered, closed contour pixel set.

[0066] The contour pixel set is traversed to determine its extreme points distributed on the two-dimensional plane, and the minimum area rectangle that can completely surround all contour pixels is fitted according to the position of the extreme points.

[0067] The total number of pixels in each candidate dirty point area is counted as the area of ​​the dirty point.

[0068] Calculate the geometric center point of each candidate dirty point and record the position coordinates of the center point in the unified coordinate system.

[0069] The area and center coordinates of the dirty point are defined as the dirty point shape features.

[0070] S4. Based on a unified coordinate system, the coordinates of the geometric features of the dots in the printed feature dataset are mapped and matched with the printing plate feature dataset, the dot enlargement rate of each corresponding dot is calculated one by one, and the dirty dot coverage rate of the corresponding area is counted to generate a deviation dataset containing position information.

[0071] The specific content of mapping and matching the coordinates of the dot geometric features in the printed product feature data set with the printing plate feature data set based on the unified coordinate system is as follows: select dot or dirty dot feature units in sequence from the printed product feature data set and read their position coordinates.

[0072] Based on the established coordinate transformation relationship, the position coordinates of the above-mentioned printed product features are reversely mapped back to the coordinate system of the original printing plate image to obtain the expected printing plate coordinate position.

[0073] Taking the expected printing plate coordinate position as the center, all feature units within the preset geometric tolerance range are searched in the printing plate feature dataset.

[0074] It should be noted that the preset geometric tolerance range is mainly determined by statistically analyzing a large number of known correct matching pairs to determine a tolerance value that can cover more than 99% of deviations. Another way is to obtain it by theoretically estimating the total error of the entire system chain.

[0075] If no printing plate feature unit is found within the tolerance range, it is determined that the printed product feature has no corresponding root cause on the printing plate, and it is recorded as an abnormal state of missing printing plate.

[0076] If multiple printing plate feature units are found within the tolerance range, it is determined that the types of these units are consistent with the current printed product feature type.

[0077] If the types are consistent, the printing plate feature unit with the closest position is selected as the corresponding feature that matches successfully, and a mapping relationship is established.

[0078] If the types are inconsistent, it is determined that the printed feature has no corresponding source of the same type on the printing plate, and it is recorded as an abnormal state of type inconsistency.

[0079] Through mapping and matching under a unified coordinate system, the connection channel between the printed product dots and the printing plate dots is opened up. This is not only the basis for calculating the dot gain rate, but also the core link for identifying dot anomalies and determining the root cause of defects.

[0080] The steps for calculating the dot enlargement ratio are as follows: based on the mapping matching result, a dot feature unit pair with a corresponding relationship established in the printing plate feature data set and the printed product feature data set is obtained.

[0081] For each pair of matching dot units, the area parameter of the dot is extracted from the printing plate feature data set, and the area parameter of the corresponding dot is extracted from the printed product feature data set.

[0082] The area difference value is obtained by subtracting the area parameter of the printed dot from the area parameter of the corresponding printing plate dot, and the dot enlargement ratio value is obtained by dividing the area difference value by the area parameter of the printing plate dot.

[0083] The dot magnification value is associated with the corresponding dot position information and dot identification information.

[0084] The standard deviation of the expansion rate of network points in each area is calculated to generate statistical indicators describing the expansion of network points in each area.

[0085] If the dot gain ratio is too low, the printed product may appear lighter in color and lose gradations, meaning that the dots in the dark areas are less saturated. If the dot gain ratio is too high, the printed product may appear too dark and blurred, meaning that the dots in the bright areas merge. Only by controlling the dot gain ratio within a reasonable range can the printed product be consistent with the digital original. Therefore, calculating the dot gain ratio is key to transforming subjective visual quality into objective data, providing a standardized basis for quality assessment.

[0086] The steps for calculating the dirty spot coverage are as follows: according to the physical layout characteristics of the printing plate and the printed product, the effective area is divided into a number of independent areas with clear boundaries.

[0087] It should be noted that the physical layout characteristics of the printing plate and printed matter refer to the physical size and shape of the effective printing area.

[0088] The effective printing area of ​​the printing plate in a unified coordinate system is used as the division object, the boundary of the effective printing area of ​​the printing plate is used as the division range, and the conventional rectangular grid in the industry is used as the standard unit of the independent area.

[0089] The characteristic data of all dirty spots on printed products in each independent area and the mapping relationship status of dirty spots on the printing plates are obtained to distinguish between valid dirty spots with corresponding root causes on the printing plates and abnormal dirty spots without corresponding root causes.

[0090] It should be noted that if there is dirt outside the effective area or on the edge of the printing plate, it will be transferred to the corresponding position of the printed product synchronously with the ink during the printing process, forming dirty spots with origins in the printing plate. If the printing plate itself is not dirty, the dirty spots that appear on the printed product can only be exogenous pollution introduced during the printing process, such as paper dust, impurities on the printing press roller, ink impurities, etc. Such dirty spots have no corresponding source on the printing plate.

[0091] Dirt spots with corresponding origins to printing plates are identified as valid dirt spots and included in the coverage statistics, while dirt spots without corresponding origins to printing plates are marked as abnormal dirt spots and recorded separately.

[0092] It should be noted that effective dirty spots are caused by dirty printing plates and are a direct reflection of quality defects in the CTP platemaking process. Their coverage directly reflects the cleanliness level of the printing plate and is a core indicator of platemaking quality control. It must be included in coverage statistics to quantitatively evaluate printing plate problems.

[0093] Abnormal dirty spots are caused by external pollution in the printing process, such as paper dust and impurities in the printing machine. They have nothing to do with the quality of the printing plate. If they are included in the coverage statistics, the statistical results will be distorted and interfere with the accurate assessment of the plate making quality.

[0094] The number of valid dirty points in each independent area is counted and combined with the area parameter of the area to obtain the dirty point coverage of the area.

[0095] The pollution degree of the dirty points in each area is determined according to the coverage rate of the dirty points in the area.

[0096] Specifically, if the coverage rate of dirty points in a certain area is less than a preset low threshold, such as 0.1%, it is judged to be slightly polluted. If the coverage rate is between the preset low threshold and the preset high threshold, where the preset low threshold can be 0.1% and the preset high threshold can be 1%, it is judged to be moderately polluted. If the coverage rate is greater than the preset high threshold, such as 1%, it is judged to be severely polluted.

[0097] Output the dirty point coverage value and pollution degree level of each area to form a complete regional dirty point coverage result.

[0098] The dirty spot coverage rate is converted from qualitative description to quantitative data, providing a standardized and replicable basis for determining the degree of pollution, avoiding subjective judgment errors, and ensuring uniform dirty spot pollution assessment standards for different regions and batches.

[0099] The specific content of the deviation data set is as follows: the enlargement ratio value and corresponding position coordinates of each dot obtained through dot mapping matching and enlargement ratio calculation.

[0100] The dirty point coverage rate and regional location identification of each area are obtained through dirty point mapping matching and coverage statistics.

[0101] Forming a deviation data set containing position information is the only data source for determining the root cause of defects. By analyzing the numerical characteristics and spatial distribution patterns of the data, it is possible to accurately distinguish whether the root cause of the defect is a printing plate problem or a printing process problem, ultimately providing structured and traceable data support for quality report generation.

[0102] S5. Generate a quality defect root cause quality report based on the deviation data set combined with the combination logic of feature position relationship and feature type.

[0103] The specific contents of the quality report on the root causes of quality defects are as follows: extracting the dot enlargement rate and corresponding position of each network dot, the dirty dot coverage rate of each area and the area position information from the deviation data set, and dividing it into network dot deviation data and dirty dot deviation data according to the feature type.

[0104] If the dot gain rates of most areas in a certain region are similar and higher than those of other areas, it is determined that the excessive dot gain is caused by abnormal printing pressure or ink volume.

[0105] It should be noted that if the printing pressure in a certain area is too high, the contact pressure between the printing plate and the rubber blanket, and between the rubber blanket and the substrate will increase overall, resulting in an increase in the ink transfer amount of all dots in the area, and then the dot area will generally expand, and the degree of expansion will tend to be similar due to the uniformity of pressure.

[0106] If too much ink is supplied to a certain area, the ink will spread excessively at the edge of the dots, causing all dots in the area to have enlarged edges. And because the ink supply system has consistent ink supply to the same area, the dot expansion rate will be generally high and the values ​​will be close.

[0107] If the difference between the gain rate of a single dot and the average gain rate of the surrounding dots exceeds the set deviation threshold, and the gain rates of the surrounding dots are normal, it is determined that the corresponding printing plate area is worn or has ink accumulation.

[0108] It should be noted that the deviation threshold is often determined through statistical process control methods.

[0109] If the difference in the average expansion rate of the surrounding dots exceeds the set deviation threshold, it means that the behavior of the dot deviates significantly from the normal trend of the area to which it directly belongs. In statistical terms, it is an outlier.

[0110] Systemic factors that cause dot gain, such as printing pressure and ink flow, typically affect a specific area. If a dot's growth rate is significantly higher than that of its immediate neighbors, this suggests that systemic pressure or ink volume settings are not the primary cause of the anomaly. Otherwise, the surrounding dots should exhibit similar anomalies.

[0111] If the coverage rate of dirty spots in a certain area of ​​the printed product is high and similar dirty spots exist in the corresponding printing plate area, it is determined that the dirty spots on the printed product are caused by the dirty printing plate.

[0112] It's important to note that if a certain area of ​​the plate is contaminated, the printing pressure will cause the contaminant to absorb ink along with the dots and transfer it to the same spatial location on the printed product, creating a spot on the printed product with the same characteristics as the plate contaminant. Conversely, if the plate is clean, the contaminant in that area of ​​the printed product can only be introduced from external sources during the printing process, such as paper dust or equipment impurities, rather than from transfer from the plate.

[0113] If the dirty spots on the printed product have no corresponding printing plate and are concentrated at the edge of the image, it is determined to be contamination by external impurities during the printing process.

[0114] It should be noted that if there is no corresponding printing plate, it means that the dirty spot has no homologous characteristics on the printing plate, which directly eliminates the root cause of the printing plate dirt transfer and proves that it is a new non-printing plate root defect added during the printing process.

[0115] The parameters involved in the above formula are all dimensionless and calculated numerically. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0116] The above embodiments may be implemented in whole or in part through software, hardware, firmware or any other combination. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0117] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0118] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0119] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0120] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A CTP platemaking quality detection and analysis method based on image recognition, characterized by: The following steps are involved: S1. Using a high-resolution image acquisition device, synchronously acquire an image of a manufactured CTP printing plate and an image of a printed product produced by the printing plate; S2. After pre-processing the collected printing plate and printed matter images, a unified coordinate system of the printing plate image and the printed matter image is established based on the physical correspondence between the printing plate and the printed matter; S3. Identify and segment single dot regions and dirty dot regions on the two images respectively, and extract dot geometric features and dirty dot shape features to form a printing plate feature dataset and a printed product feature dataset; S4. Mapping and matching the coordinates of the dot geometric features in the printed product feature dataset with the printing plate feature dataset based on a unified coordinate system, calculating the dot gain rate of each corresponding dot one by one, and calculating the dirty dot coverage rate of the corresponding area to generate a deviation dataset containing position information; S5. Generate a quality defect root cause quality report based on the deviation data set combined with the combination logic of feature position relationship and feature type.

2. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The specific steps of establishing a unified coordinate system for the printing plate image and the printed product image are as follows: When making plates, mark the geometric center and corner points outside the effective printing area of ​​the printing plate as physical reference marks; identifying a plurality of non-collinear physical fiducial marks in the pre-processed printing plate image and the printed product image respectively; Obtaining the pixel coordinates of each physical reference mark in the printing plate image and the corresponding pixel coordinates in the printed product image; Based on the two sets of pixel coordinates of the physical reference marks in the printing plate image and the printed product image, a coordinate transformation matrix is ​​constructed from the printed product image coordinates to the printing plate image coordinates, and the pixel coordinates of all feature points in the printed product image are uniformly transformed into the coordinate system of the printing plate image; The vertices of the plate image are selected as the origin of the unified coordinate system.

3. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The specific steps of segmenting the single dot area and the dirty dot area are as follows: Perform connected region detection on the preprocessed printing plate image and printed product image, and calculate the mean grayscale value of each connected region pixel; If the grayscale mean falls within the preset dot grayscale value interval, the connected area is marked as a dot area and divided into independent dot units; The pre-processed plate image and printed product image are converted into grayscale images and then binarized; The area with a grayscale value greater than a preset grayscale threshold after binarization is divided into a foreground pixel area, otherwise it is divided into a background pixel area; If the area of ​​the foreground pixel region exceeds the noise threshold, it is marked as a dirty point region, and the dirty point region is segmented to generate independent dirty point units.

4. The CTP platemaking quality detection and analysis method based on image recognition according to claim 3, characterized in that: The steps of extracting the geometric features of the mesh points are as follows: Mark the connected area whose grayscale mean falls within the preset dot grayscale value interval as the target connected area; Count the total number of pixels included in the target connected domain to determine the area parameters of the dots; Calculate the geometric center of the connected domain through the pixel coordinates of the contour boundary, and record the position coordinates of the center in the unified coordinate system; The area parameters and center coordinates of the dots are defined as the dot geometric features.

5. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The steps of extracting the dirty point shape features are as follows: Scan the binary image to identify the continuous region formed by all foreground pixels, and regard each continuous region as a candidate dirty point; Perform contour tracing on each candidate dirty point to obtain its boundary contour pixel set; Traverse the contour pixel set to determine its extreme points distributed on the two-dimensional plane, and fit the minimum area rectangle that can completely surround all contour pixels based on the position of the extreme points; Count the total number of pixels in each candidate dirty point area as the area of ​​the dirty point; Calculate the geometric center point of each candidate dirty point and record the position coordinates of the center point in the unified coordinate system; The area and center coordinates of the dirty point are defined as the dirty point shape features.

6. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The specific contents of mapping and matching the coordinates of the dot geometric features in the printed product feature data set with the printing plate feature data set based on the unified coordinate system are as follows: Selecting dot or dirty dot feature units in sequence from the printed feature data set and reading their position coordinates; Based on the established coordinate transformation relationship, the position coordinates of the printed feature are reversely mapped back to the coordinate system of the original printing plate image to obtain the expected printing plate coordinate position; Taking the expected printing plate coordinate position as the center, searching for all feature units within a preset geometric tolerance range in the printing plate feature data set; If no printing plate feature unit is found within the tolerance range, it is determined that the printed feature has no corresponding root cause on the printing plate, and it is recorded as an abnormal state of missing printing plate; If multiple printing plate feature units are found within the tolerance range, it is determined whether the types of these units are consistent with the current printed product feature type; If the types are consistent, the printing plate feature unit with the closest position is selected as the corresponding feature that matches successfully, and a mapping relationship is established; If the types are inconsistent, it is determined that the printed feature has no corresponding source of the same type on the printing plate, and it is recorded as an abnormal state of type inconsistency.

7. The CTP platemaking quality detection and analysis method based on image recognition according to claim 4, characterized in that: The steps for calculating the dot gain rate are as follows: Acquire, based on the mapping matching result, a dot feature unit pair that has established a corresponding relationship in the printing plate feature dataset and the printed product feature dataset; For each pair of matching dot units, the area parameter of the dot is extracted from the printing plate feature data set, and the area parameter of the corresponding dot is extracted from the printed product feature data set; The area parameter of the printed dot is subtracted from the area parameter of the corresponding printing plate dot to obtain an area difference value, and the area difference value is divided by the area parameter of the printing plate dot to obtain a dot enlargement ratio value; Associating the dot gain ratio value with the corresponding dot location information and dot identification information; The standard deviation of the expansion rate of network points in each area is calculated to generate statistical indicators describing the expansion of network points in each area.

8. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The steps for calculating the dirty point coverage are as follows: According to the physical layout characteristics of the printing plate and printed matter, the effective area is divided into several independent areas with clear boundaries; Obtain the characteristic data of all dirty spots on printed products in each independent area and the mapping relationship status of dirty spots on the printing plates, and distinguish between valid dirty spots with corresponding root causes on the printing plates and abnormal dirty spots without corresponding root causes; Dirt spots with corresponding origins to the printing plate are identified as valid dirt spots and included in the coverage statistics, while dirt spots without corresponding origins to the printing plate are marked as abnormal dirt spots and recorded separately; The number of valid dirty points in each independent area is counted and combined with the area parameter of the area to obtain the dirty point coverage rate of the area; Determine the pollution level of the dirty spots in each area based on the coverage rate of the dirty spots in the area; Output the dirty point coverage value and pollution degree level of each area to form a complete regional dirty point coverage result.

9. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The specific contents of the deviation data set are as follows: The enlargement ratio value and corresponding position coordinates of each dot obtained through dot mapping matching and enlargement ratio calculation; The dirty point coverage rate and regional location identification of each area are obtained through dirty point mapping matching and coverage statistics.

10. The CTP platemaking quality detection and analysis method based on image recognition according to claim 1, characterized in that: The specific contents of the quality defect root cause quality report are as follows: Extract the dot enlargement rate and corresponding position of each network point, the dirty point coverage rate of each area and the regional location information from the deviation data set, and divide it into network dot deviation data and dirty point deviation data according to feature type; If the dot gain rates of most areas in a certain region are similar and higher than those of other areas, it is determined that the excessive dot gain is caused by abnormal printing pressure or ink volume; If the difference between the gain rate of a single dot and the average gain rate of the surrounding dots exceeds the set deviation threshold, and the gain rates of the surrounding dots are normal, it is determined that the corresponding printing plate area is worn or has ink accumulation; If the coverage rate of dirty spots in a certain area of ​​the printed product is high and similar dirty spots exist in the corresponding printing plate area, it is determined that the dirty spots on the printed product are caused by the dirty printing plate; If the dirty spots on the printed product have no corresponding printing plate and are concentrated at the edge of the image, it is determined to be contamination by external impurities during the printing process.

Images