Image recognition-based color-coated sheet design and color pattern printing precision detection method

Abstract

The invention discloses a color-coated sheet color pattern printing precision detection method based on image recognition, and relates to the technical field of printing precision detection. The method comprises the following steps: acquiring printing production multi-dimensional influence data, and determining whether a color-coated sheet design and color pattern printing environment is normal or not based on the printing production multi-dimensional influence data: if the color-coated sheet design and color pattern printing environment is abnormal, stopping printing; if the color-coated sheet color pattern printing environment is normal, the printed color-coated sheet is obtained, a color-coated sheet image is extracted and preprocessed, and a preprocessed color-coated sheet image is obtained; performing pattern processing on the color-coated sheet image to obtain a color-coated sheet color pattern, and obtaining a printing precision error coefficient; the printing precision error threshold value is determined based on the printing production multi-dimensional influence data, the printing precision is evaluated in combination with the printing precision error coefficient, and the problem that an existing color-coated sheet color and pattern printing precision detection method is difficult to comprehensively and accurately detect the color-coated sheet color and pattern printing precision is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of printing precision detection, and in particular to a method for detecting the printing precision of color patterns on color-coated plates based on image recognition. Background Art

[0002] There are various methods for inspecting the printing accuracy of color and pattern patterns on color-coated steel sheets, each with its own characteristics and suitability for different scenarios and accuracy requirements. Common methods include visual inspection, dimensional measurement, optical inspection, and image analysis. Image analysis methods include those based on scanning imaging and software analysis, and those based on artificial intelligence image recognition.

[0003] Scanning, imaging, and software analysis involve scanning or photographing the printed pattern on the color-coated board using a scanner or high-resolution camera to obtain a digital image. This image is then processed and analyzed using specialized image analysis software. The software automatically identifies pattern features and measures various parameters, such as line position accuracy, pattern repeatability, and color uniformity. The software then compares the image with a pre-set standard image or design file, quickly and accurately calculating deviations and generating a detailed inspection report.

[0004] AI-based image recognition testing utilizes a deep learning algorithm trained on a large number of images of printed patterns on color-coated sheet metal. This allows the computer model to learn the characteristics of normal and defective patterns. In actual testing, the model can automatically identify various defects and deviations within the pattern, and can even classify and count different types of defects.

[0005] The existing printing accuracy detection method based on image analysis has the problem that it is difficult to comprehensively and accurately detect the printing accuracy of color-coated plate patterns by combining printing equipment, workshop environment, coating characteristics and color-coated plate characteristics. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention provides a method for detecting the printing accuracy of color-coated plate patterns based on image recognition, which solves the problem that the existing method for detecting the printing accuracy of color-coated plate patterns is difficult to comprehensively and accurately detect the printing accuracy of color-coated plate patterns.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for detecting the printing accuracy of color-coated plate patterns based on image recognition, comprising the following steps: obtaining multi-dimensional impact data of printing production, and determining whether the printing environment of the color-coated plate patterns is normal based on the multi-dimensional impact data generated by printing: if the printing environment of the color-coated plate patterns is abnormal, stopping printing; if the printing environment of the color-coated plate patterns is normal, obtaining the printed color-coated plate, extracting the color-coated plate image for preprocessing, and obtaining the preprocessed color-coated plate image; performing pattern processing on the color-coated plate image to obtain the color-coated plate patterns, and obtaining the printing accuracy error coefficient; determining the printing accuracy error threshold based on the multi-dimensional impact data of printing production, and evaluating the printing accuracy in combination with the printing accuracy error coefficient.

[0008] Furthermore, the multi-dimensional influencing data of printing production includes printing equipment operation data, printing workshop environmental parameter data, coating characteristic parameter data and color-coated board characteristic parameter data.

[0009] Furthermore, based on the generation of multi-dimensional impact data based on printing, whether the printing environment of the color-coated plate pattern is normal is determined, including the following steps: using the printing equipment operation data as the input of the support vector machine model to obtain the operation status of the printing equipment. If the operation status of the printing equipment is abnormal, the color-coated plate pattern printing environment is abnormal; if the printing equipment operation status is normal, using the printing workshop environmental parameter data as the input of the environmental factor judgment model constructed based on the convolutional neural network to obtain the printing workshop environmental status. If the printing workshop environmental status is abnormal, the color-coated plate pattern printing environment is abnormal; if the printing workshop environmental status is normal, using the coating characteristic parameter data and the color-coated plate characteristic parameter data as the input of the random forest model to obtain the coating characteristics and the color-coated plate characteristic status. If the coating characteristics and the color-coated plate characteristic status are abnormal, the color-coated plate pattern printing environment is abnormal; if the coating characteristics and the color-coated plate characteristic status are normal, the color-coated plate pattern printing environment is normal.

[0010] Furthermore, the color-coated plate image is pattern-processed to obtain the color-coated plate pattern, including the following steps: performing edge detection on the color-coated plate image to obtain an edge contour image; performing corner detection on the edge contour image based on a corner detection algorithm to obtain a set of corner coordinates; performing threshold segmentation based on the edge contour image and the set of corner coordinates to obtain the color-coated plate pattern.

[0011] Furthermore, edge detection is performed on the color-coated plate image to obtain an edge contour image, which includes the following steps: smoothing the color-coated plate image and obtaining a gradient image based on the Sobel operator; performing non-maximum suppression on the gradient image to obtain a refined edge image; obtaining an initial dual-threshold detection interval stored in a database; obtaining a printing workshop environment-dual-threshold correction coefficient mapping database stored in the database, and retrieving the corresponding dual-threshold correction coefficient based on the current printing workshop environment parameter data; correcting the initial dual-threshold detection interval based on the dual-threshold correction coefficient to obtain a corrected dual-threshold detection interval; obtaining the pixel value of each pixel point in the refined edge image: if the pixel value of a certain pixel point is greater than the maximum value in the corrected dual-threshold detection interval, the pixel point is identified as an edge pixel point; if the pixel value of a certain pixel point is less than the minimum value in the corrected dual-threshold detection interval, the pixel point is identified as a non-edge pixel point; if the pixel value of a certain pixel point is between the corrected dual-threshold detection interval, determining whether the pixel point is connected to the edge pixel point, if connected, it is identified as an edge pixel point, and if not connected, it is identified as a non-edge pixel point; connecting all edge pixels to obtain an edge contour image.

[0012] Furthermore, threshold segmentation is performed based on the edge contour image and the corner point coordinate set to obtain the color pattern of the color-coated plate, including the following steps: obtaining the pixel grayscale values of all pixels in the edge contour image, drawing the grayscale value distribution, determining the most concentrated grayscale interval, and defining the most concentrated grayscale interval as the initial threshold segmentation range; traversing the pixel grayscale values of the domain pixels of each corner point in the corner point coordinate set to obtain the mean and standard deviation, if the mean is lower than the mean of the pixel grayscale values of all pixels in the edge contour image and the standard deviation is greater than the set standard deviation threshold, then obtaining the pollutant data, and correcting the initial threshold segmentation range based on the pollutant data to obtain the corrected threshold segmentation range, otherwise the initial threshold segmentation range is not corrected; if the pixel grayscale value of a certain pixel point in the color-coated plate image is within the corrected threshold segmentation range or the uncorrected initial threshold segmentation range, then the pixel point is recorded as the pattern area, otherwise it is recorded as the background area to obtain the color pattern of the color-coated plate.

[0013] Furthermore, obtaining the printing accuracy error coefficient includes the following steps: obtaining the color standard pattern of the color coated plate stored in the database, aligning the color standard pattern of the color coated plate with the color standard pattern of the color coated plate; obtaining the size error Cw, position error Ww and color error Yw; and obtaining the printing accuracy error coefficient Yx based on the error accuracy evaluation formula:

[0014] Yx=ω1*Cw+ω2*Ww+ω3*Yw;

[0015] Among them, ω1 is the weight factor of Cw, ω2 is the weight factor of Ww, and ω3 is the weight factor of Yw.

[0016] Furthermore, a printing accuracy error threshold is determined based on the multidimensional influencing data of printing production, including the following steps: using the printing equipment operation data and the printing workshop environmental parameter data as inputs of the decision tree model to obtain the size theoretical error Clw and the position theoretical error Wlw; using the coating characteristic parameter data and the printing workshop environmental parameter data as inputs of the multivariate linear regression model to obtain the first color theoretical error fYlw; using the color-coated board characteristic parameter data and the printing workshop environmental parameter data as inputs of the random forest regression model to obtain the second color theoretical error sYlw; and obtaining the printing accuracy error threshold Yxy based on the size theoretical error Clw, the position theoretical error Wlw, the first color theoretical error fYlw and the second color theoretical error sYlw:

[0017]

[0018] Among them, ω1′ is the weight factor of Clw, ω2′ is the weight factor of Wlw, and ω3′ is The weight factor of .

[0019] Furthermore, the printing accuracy is evaluated in combination with the printing accuracy error coefficient, including the following steps: if the printing accuracy error coefficient is less than the printing accuracy error threshold, the accuracy detection result is normal; if the printing accuracy error coefficient is not less than the printing accuracy error threshold, the accuracy detection result is abnormal, and the multi-dimensional influencing data of printing production is optimized and regulated.

[0020] Furthermore, the multi-dimensional influencing data of printing production is optimized and regulated, including the following steps: establishing a size and position error-equipment and environment data mapping database based on the input and output of the decision tree model, establishing a first color error-coating and environment data mapping database based on the input and output of the random forest regression model, and establishing a second color error-color-coated board and environment data mapping database based on the input and output of the random forest regression model; obtaining the error allowable thresholds stored in the database, including the size error allowable threshold, the position error allowable threshold, the first color error allowable threshold, and the second color error allowable threshold;

[0021] Screening the combination of multi-dimensional impact data on printing production to be evaluated from the database of size and position error-equipment and environment data mapping, the first database of color error-coating and environment data mapping, and the second database of color error-color-coated board and environment data mapping;

[0022] Obtain corresponding comprehensive error values dYx for each combination of multi-dimensional impact data of selected printing production to be evaluated; determine the cost-effectiveness Xyb and implementation feasibility score Pf for each combination of multi-dimensional impact data of selected printing production to be evaluated;

[0023] Determine each evaluation coefficient px;

[0024]

[0025] Among them, β1 is The weight factor of β2 is the weight factor of Xyb, and β3 is The weight factor, e is a natural constant;

[0026] The combination of multi-dimensional impact data of printing production to be evaluated corresponding to the largest evaluation coefficient is used as the adjustment target, and the multi-dimensional impact data of printing production is optimized and regulated.

[0027] The present invention has the following beneficial effects:

[0028] This method for detecting the printing accuracy of color-coated plate patterns based on image recognition comprehensively considers the impact of multiple factors such as equipment, environment, coatings and color-coated plates on printing accuracy, promptly detects abnormal conditions in the printing process, stops printing when the printing environment is abnormal to avoid producing unqualified products, and can also make quantitative judgments on printing accuracy based on the evaluation results, providing a basis for subsequent optimization and regulation, improving the printing quality of color-coated plates, reducing production costs, and solving the problem of difficulty in comprehensively and accurately detecting the printing accuracy of color-coated plate patterns in existing technologies.

[0029] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The present invention is a flow chart of the method for detecting the printing accuracy of color-coated plate patterns based on image recognition.

[0031] Figure 2 The present invention is a flow chart for evaluating printing accuracy of a method for detecting printing accuracy of color-coated plate patterns based on image recognition. DETAILED DESCRIPTION

[0032] See also Figure 1 The embodiment of the present invention provides a technical solution: a method for detecting the printing accuracy of color-coated plate patterns based on image recognition, comprising the following steps: obtaining multi-dimensional impact data of printing production, and determining whether the color-coated plate pattern printing environment is normal based on the multi-dimensional impact data generated by printing:

[0033] The multi-dimensional influencing data of printing production include printing equipment operation data (including printing roller pressure, printing speed, roller diameter deviation, transmission system accuracy, positioning device error), printing workshop environmental parameter data (temperature and humidity, light intensity, air quality, vibration, airflow), coating characteristic parameter data (pigment particle size, pigment concentration, drying speed) and color-coated board characteristic parameter data (surface roughness, material adsorption to coating, coating thickness).

[0034] Using printing equipment operating data as input to the support vector machine model, the operating status of the printing equipment is determined. Stable operation of the printing equipment is the basis for ensuring printing accuracy. By analyzing equipment operating data such as printing roller pressure and printing speed through the support vector machine model, it can quickly and accurately determine whether the equipment is in normal working condition.

[0035] If the printing equipment is operating abnormally, the printing environment for the color-coated board pattern is abnormal. If the printing equipment is operating normally, the printing workshop environmental parameter data is used as input to the environmental factor judgment model built based on a convolutional neural network to obtain the printing workshop environmental status. The workshop environment has a significant impact on printing quality. Changes in environmental parameters such as temperature, humidity, and light intensity can affect the drying speed of the coating and the physical properties of the color-coated board. The model built using the convolutional neural network analyzes these parameters and can promptly detect environmental anomalies.

[0036] If the printing room environment is abnormal, then the printing environment for the color-coated board pattern is abnormal. If the printing room environment is normal, the coating and board characteristic parameter data are used as inputs to the random forest model to obtain the coating and board characteristics. The characteristics of the coating and board themselves are key factors in determining printing accuracy. Parameters such as pigment particle size and concentration, as well as the surface roughness of the board and its adsorption to the coating, all affect the rendering of the pattern. Using the random forest model to analyze these characteristic parameters can effectively determine whether they are normal.

[0037] If the coating properties and the color-coated plate property status are abnormal, the color-coated plate pattern printing environment is abnormal; if the coating properties and the color-coated plate property status are normal, the color-coated plate pattern printing environment is normal.

[0038] If the printing environment of the color-coated plate pattern is abnormal, printing is stopped; if the printing environment of the color-coated plate pattern is normal, the printed color-coated plate is obtained, the color-coated plate image is extracted for preprocessing, and the preprocessed color-coated plate image is obtained; the actual collected color-coated plate image has problems such as noise interference and uneven lighting, which will affect the subsequent pattern analysis and feature extraction. Preprocessing operations (such as denoising, grayscale, histogram equalization, etc.) can effectively improve image quality, remove noise, enhance image contrast and clarity, make the pattern features in the image more obvious, and provide high-quality basic images for subsequent edge detection, corner detection and other operations, thereby improving the accuracy and reliability of the entire detection process.

[0039] Perform edge detection on color-coated steel plate images to obtain edge contour images. Edges are important features of patterns in images, containing information about their shape and structure. Edge detection can accurately locate the boundaries of patterns and separate them from the background.

[0040] After smoothing the color-coated plate image, a gradient image is obtained based on the Sobel operator. The Sobel operator can calculate the gradient of the image in the horizontal and vertical directions, highlighting the areas with obvious grayscale changes, that is, potential pattern edges, and providing preliminary gradient information for subsequent accurate edge recognition.

[0041] Non-maximum suppression is performed on the gradient image to obtain a refined edge image. Non-maximum suppression only retains the local maximum pixel points in the gradient direction, removes other redundant pixels, and refines the edges so that the edges more accurately reflect the pattern boundaries, reducing the error in edge position judgment in subsequent processing and improving the accuracy of edge detection.

[0042] Obtain the initial dual-threshold detection interval stored in the database; obtain the printing workshop environment-dual-threshold correction coefficient mapping database stored in the database, and retrieve the corresponding dual-threshold correction coefficient based on the current printing workshop environmental parameter data; correct the initial dual-threshold detection interval based on the dual-threshold correction coefficient (multiply the coefficient by the interval endpoint value) to obtain the corrected dual-threshold detection interval. Different printing workshop environments (temperature, humidity, lighting, etc.) can affect image quality and edge characteristics. Obtaining the initial dual-threshold detection interval from the database is the basic basis for edge judgment. In combination with the current printing workshop environmental parameters, retrieving the coefficient correction threshold interval from the "printing workshop environment-dual-threshold correction coefficient mapping database" can adapt to different environmental changes and make edge detection more accurate.

[0043] Obtain the pixel value of each pixel in the refined edge image: if the pixel value of a certain pixel is greater than the maximum value in the corrected dual-threshold detection interval, the pixel is identified as an edge pixel; if the pixel value of a certain pixel is less than the minimum value in the corrected dual-threshold detection interval, the pixel is identified as a non-edge pixel; if the pixel value of a certain pixel is between the corrected dual-threshold detection interval, determine whether the pixel is connected to the edge pixel, if connected, it is identified as an edge pixel, if not connected, it is identified as a non-edge pixel; connect all edge pixels to obtain an edge contour image.

[0044] Corner detection is performed on edge contour images using a corner detection algorithm (Harris corner detection or SIFT feature point detection) to obtain a set of corner coordinates. Corner points are distinct features in an image, typically located at pattern corners or edge intersections. Corner points have unique geometric characteristics and stability, providing important reference points for pattern matching and positioning. By detecting corner point coordinates, key locations in the pattern can be determined, further improving detection accuracy.

[0045] Threshold segmentation is performed based on the edge outline image and the set of corner point coordinates to obtain the color pattern of the color-coated steel sheet. Threshold segmentation is a simple and effective image segmentation method that divides the image into pattern and background regions by comparing the grayscale or color values of pixels in the image with a set threshold. Combining the edge outline image and the set of corner point coordinates for threshold segmentation can more accurately determine the pattern range, remove background noise interference, and obtain a clear color pattern of the color-coated steel sheet.

[0046] Obtain the grayscale values of all pixels within the edge contour image, plot the grayscale distribution, and identify the interval with the highest grayscale concentration. This interval is then used as the initial threshold segmentation range. Obtaining the grayscale values of all pixels and plotting their distribution provides a visual representation of the image's grayscale concentration. Using the interval with the highest grayscale concentration as the initial threshold effectively distinguishes pattern from background. Because grayscale values within a patterned area are typically similar and concentrated, this method can provide a preliminary delineation of the pattern's extent.

[0047] Traverse the pixel grayscale values of the domain pixels of each corner point in the corner point coordinate set to obtain the mean and standard deviation. If the mean is lower than the mean of the pixel grayscale values of all pixels in the edge contour image and the standard deviation is greater than the set standard deviation threshold, obtain the pollutant data, and correct the initial threshold segmentation range based on the pollutant data to obtain the corrected threshold segmentation range. Otherwise, do not correct the initial threshold segmentation range (obtain the pollutant-threshold segmentation range correction coefficient mapping database stored in the database, retrieve the corresponding threshold segmentation range correction coefficient based on the current pollutant data, and correct the initial threshold segmentation range based on the threshold segmentation range correction coefficient, and multiply the coefficient by the interval endpoint value); based on this, obtain pollutant data and correct the threshold range, which can adapt to complex image conditions and more accurately define the pattern area.

[0048] If the grayscale value of a certain pixel in the color-coated plate image is within the corrected threshold segmentation range or the uncorrected initial threshold segmentation range, the pixel is recorded as the pattern area, otherwise it is recorded as the background area, and the color pattern of the color-coated plate is obtained.

[0049] Perform pattern processing on the color-coated plate image to obtain the color pattern of the color-coated plate and obtain the printing accuracy error coefficient;

[0050] Obtaining the printing accuracy error coefficient includes the following steps: obtaining the color standard pattern of the color coated plate stored in the database, aligning the color standard pattern of the color coated plate with the color standard pattern of the color coated plate; obtaining the size error Cw, position error Ww and color error Yw; and obtaining the printing accuracy error coefficient Yx based on the error accuracy evaluation formula:

[0051] Yx=ω1*Cw+ω2*Ww+ω3*Yw;

[0052] Among them, ω1 is the weight factor of Cw, ω2 is the weight factor of Ww, and ω3 is the weight factor of Yw.

[0053] Size error reflects the difference in size between the pattern and the standard, position error reflects the deviation of the pattern in space, and color error shows the degree of deviation of the pattern color from the standard. Comprehensively capturing these three types of error data can measure print quality from multiple dimensions.

[0054] The printing accuracy error threshold is determined based on the multi-dimensional influencing data of printing production, and the printing accuracy is evaluated in combination with the printing accuracy error coefficient.

[0055] Printing equipment operating data and pressroom environmental parameter data are used as inputs to a decision tree model to determine the theoretical size error Clw and the theoretical position error Wlw. Printing equipment operating data (such as printing roller pressure and speed) and pressroom environmental parameters (such as temperature, humidity, and vibration) significantly influence pattern size and position accuracy. The decision tree model predicts the theoretical size and position error ranges based on historical data and equipment and environmental parameters. The model can exploit the relationships between these complex data.

[0056] The first theoretical color error, fYlw, is derived by using coating characteristic parameter data and pressroom environmental parameter data as inputs to a multiple linear regression model. Coating characteristics (such as pigment particle size and concentration) and environmental parameters jointly influence the color of printed patterns. The multiple linear regression model quantitatively analyzes their relationship with color error and calculates the first theoretical color error based on the current coating and environmental data.

[0057] The random forest regression model uses the characteristic parameter data of the color-coated board and the environmental parameters of the printing workshop as input to obtain the second color theoretical error sYlw. The characteristics of the color-coated board (surface roughness, adsorption, etc.) also affect the color effect. The random forest regression model can comprehensively consider the characteristics of the color-coated board and the environmental parameters to further analyze the color error.

[0058] The printing accuracy error threshold Yxy is obtained based on the size theoretical error Clw, the position theoretical error Wlw, the first color theoretical error fYlw and the second color theoretical error sYlw:

[0059]

[0060] Among them, ω1′ is the weight factor of Clw, ω2′ is the weight factor of Wlw, and ω3′ is The weight factor of .

[0061] like Figure 2As shown, if the printing accuracy error coefficient is less than the printing accuracy error threshold, the accuracy detection result is normal; if the printing accuracy error coefficient is not less than the printing accuracy error threshold, the accuracy detection result is abnormal, and the multi-dimensional influencing data of printing production is optimized and regulated.

[0062] A size and position error-equipment and environment data mapping database is established based on the input and output of the decision tree model, a first color error-coating and environment data mapping database is established based on the input and output of the random forest regression model, and a second color error-color-coated board and environment data mapping database is established based on the input and output of the random forest regression model; and error tolerance thresholds stored in the database are obtained, including a size error tolerance threshold, a position error tolerance threshold, a first color error tolerance threshold, and a second color error tolerance threshold.

[0063] The multi-dimensional influencing data combination of printing production to be evaluated is selected from the dimensional position error-equipment and environmental data mapping database, the first color error-coating and environmental data mapping database, and the second color error-color coated board and environmental data mapping database (each error is required to be within the corresponding error allowable threshold. Since the printing workshop environmental parameter data will be repeatedly obtained, the three sets of printing workshop environmental parameter data can be averaged); based on the input and output of different models, the dimensional position error-equipment and environmental data, the first color error-coating and environmental data, and the second color error-color coated board and environmental data mapping databases are established respectively, which helps to clearly sort out the relationship between each influencing factor and error, and provide data support for subsequent analysis.

[0064] Obtain the corresponding comprehensive error value dYx for each combination of multi-dimensional impact data on selected printing production to be evaluated (input into the decision tree model, the multivariate linear regression model, and the random forest regression model); determine the cost-effectiveness Xyb and implementation feasibility score Pf for each combination of multi-dimensional impact data on selected printing production to be evaluated (pre-establish relevant evaluation basis, and perform a comprehensive score based on experts' assessment of the operability of the combination of multi-dimensional impact data on selected printing production to be evaluated in actual production, including the difficulty of equipment adjustment (such as whether large-scale equipment modification is required), the technical requirements for environmental control (whether existing equipment can meet them), the complexity of coating formula improvement (whether special raw materials or processes are required), and the feasibility of adjusting the characteristics of color-coated board (whether it affects the production cycle, etc.));

[0065] Determine each evaluation coefficient px;

[0066]

[0067] Among them, β1 is The weight factor of β2 is the weight factor of Xyb, and β3 is The weight factor, e is a natural constant;

[0068] The combination of multi-dimensional impact data of printing production to be evaluated corresponding to the largest evaluation coefficient is used as the adjustment target, and the multi-dimensional impact data of printing production is optimized and regulated.

[0069] The evaluation coefficient is calculated according to a pre-set formula, and the combination of data to be evaluated corresponding to the maximum evaluation coefficient is used as the adjustment target. By comprehensively considering error values, cost-effectiveness, and feasibility, the optimal parameter adjustment solution can be selected, achieving precise optimization and control of the multi-dimensional data affecting printing production, effectively improving printing accuracy while balancing quality, cost, and feasibility factors in production.

[0070] An electronic device includes: a processor; and a memory, wherein computer program instructions are stored in the memory, and when the computer program instructions are executed by the processor, the processor executes the above-mentioned method for detecting the printing accuracy of color-coated plate patterns based on image recognition.

[0071] A computer-readable storage medium is used to store a program, which, when executed by a processor, implements the above-mentioned method for detecting the printing accuracy of color-coated plate patterns based on image recognition.

[0072] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0073] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0074] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0075] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0076] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0077] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for detecting the printing accuracy of color-coated plate patterns based on image recognition, characterized in that: The following steps are involved: Obtain multi-dimensional impact data of printing production, and determine whether the printing environment of color-coated plate patterns is normal based on the multi-dimensional impact data generated by printing: If the printing environment of the color-coated board pattern is abnormal, stop printing; If the printing environment of the color-coated plate pattern is normal, obtain the printed color-coated plate, extract the color-coated plate image for preprocessing, and obtain the preprocessed color-coated plate image; Perform pattern processing on the color-coated plate image to obtain the color pattern of the color-coated plate and obtain the printing accuracy error coefficient; The printing accuracy error threshold is determined based on the multi-dimensional influencing data of printing production, and the printing accuracy is evaluated in combination with the printing accuracy error coefficient.

2. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 1, characterized in that: The multi-dimensional influencing data of printing production include printing equipment operation data, printing workshop environmental parameter data, coating characteristic parameter data and color-coated board characteristic parameter data.

3. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 2, characterized in that: Determining whether the printing environment of color-coated plate patterns is normal based on the multi-dimensional impact data generated by printing includes the following steps: The printing equipment operation data is used as the input of the support vector machine model to obtain the operation status of the printing equipment. If the operation status of the printing equipment is abnormal, the printing environment of the color-coated plate pattern is abnormal. If the printing equipment is operating normally, the environmental parameter data of the printing workshop is used as the input of the environmental factor judgment model built based on the convolutional neural network to obtain the environmental status of the printing workshop. If the environmental status of the printing workshop is abnormal, the printing environment of the color-coated plate pattern is abnormal. If the printing workshop environment is normal, the coating characteristic parameter data and the color-coated plate characteristic parameter data are used as inputs to the random forest model to obtain the coating characteristics and the color-coated plate characteristic status. If the coating characteristics and the color-coated plate characteristic status are abnormal, then the color-coated plate pattern printing environment is abnormal; If the coating properties and color-coated board properties are normal, the color-coated board pattern printing environment is normal.

4. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 2, characterized in that: The color-coated plate image is subjected to pattern processing to obtain the color-coated plate pattern, comprising the following steps: Perform edge detection on the color-coated plate image to obtain the edge contour image; Perform corner detection on the edge contour image based on the corner detection algorithm to obtain a set of corner coordinates; Threshold segmentation is performed based on the edge contour image and the corner point coordinate set to obtain the color pattern of the color-coated plate.

5. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 4, characterized in that: Performing edge detection on the color-coated plate image to obtain the edge contour image includes the following steps: After smoothing the color-coated plate image, the gradient image is obtained based on the Sobel operator; Perform non-maximum suppression on the gradient image to obtain the refined edge image; Obtaining the initial dual-threshold detection interval stored in the database; Obtaining a printing workshop environment-dual threshold correction coefficient mapping database stored in a database, and retrieving corresponding dual threshold correction coefficients based on current printing workshop environment parameter data; The initial dual-threshold detection interval is corrected based on the dual-threshold correction coefficient to obtain a corrected dual-threshold detection interval; Get the pixel value of each pixel in the refined edge image: If the pixel value of a certain pixel is greater than the maximum value in the corrected dual threshold detection interval, the pixel is identified as an edge pixel; If the pixel value of a certain pixel is less than the minimum value in the corrected dual threshold detection interval, the pixel is identified as a non-edge pixel; If the pixel value of a certain pixel point is within the corrected dual threshold detection interval, it is determined whether the pixel point is connected to the edge pixel point. If so, it is identified as an edge pixel point; if not, it is identified as a non-edge pixel point. Connect all edge pixels to obtain the edge contour image.

6. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 4, characterized in that: Threshold segmentation is performed based on the edge contour image and the corner point coordinate set to obtain the color pattern of the color-coated plate, including the following steps: Obtain the pixel grayscale values of all pixels in the edge contour image, draw the grayscale value distribution, determine the most concentrated grayscale interval, and define the most concentrated grayscale interval as the initial threshold segmentation range; Traverse the pixel grayscale values of the area pixels of each corner point in the corner point coordinate set, obtain the mean and standard deviation, if the mean is lower than the mean of the pixel grayscale values of all pixels in the edge contour image and the standard deviation is greater than the set standard deviation threshold, then obtain the pollutant data, and correct the initial threshold segmentation range based on the pollutant data to obtain the corrected threshold segmentation range, otherwise do not correct the initial threshold segmentation range; If the grayscale value of a certain pixel in the color-coated plate image is within the corrected threshold segmentation range or the uncorrected initial threshold segmentation range, the pixel is recorded as the pattern area, otherwise it is recorded as the background area, and the color pattern of the color-coated plate is obtained.

7. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 1, characterized in that: Obtaining the printing accuracy error coefficient includes the following steps: Obtaining the color standard pattern of the color coated plate stored in the database, and aligning the color pattern of the color coated plate with the color standard pattern of the color coated plate; Get size error Cw, position error Ww and color error Yw; Based on the error accuracy evaluation formula, the printing accuracy error coefficient Yx is obtained: Yx=ω1*Cw+ω2*Ww+ω3*Yw; Among them, ω1 is the weight factor of Cw, ω2 is the weight factor of Ww, and ω3 is the weight factor of Yw.

8. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 2, characterized in that: Determining a printing accuracy error threshold based on multi-dimensional influencing data of printing production includes the following steps: The printing equipment operation data and the printing workshop environmental parameter data are used as the input of the decision tree model to obtain the theoretical size error Clw and the theoretical position error Wlw; The coating characteristic parameter data and the printing workshop environmental parameter data are used as the input of the multiple linear regression model to obtain the first color theory error fYlw; The characteristic parameter data of color-coated board and the environmental parameter data of printing workshop are used as the input of random forest regression model to obtain the second color theory error sYlw; The printing accuracy error threshold Yxy is obtained based on the size theoretical error Clw, the position theoretical error Wlw, the first color theoretical error fYlw and the second color theoretical error sYlw: Among them, ω1′ is the weight factor of Clw, ω2′ is the weight factor of Wlw, and ω3′ is The weight factor of .

9. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 8, characterized in that: The evaluation of printing accuracy is carried out in combination with the printing accuracy error coefficient, including the following steps: If the printing accuracy error coefficient is less than the printing accuracy error threshold, the accuracy test result is normal; If the printing accuracy error coefficient is not less than the printing accuracy error threshold, the accuracy detection result is abnormal, and the multi-dimensional influencing data of printing production is optimized and regulated.

10. The method for detecting printing accuracy of color-coated plate patterns based on image recognition according to claim 9, characterized in that: Optimizing and controlling the multi-dimensional impact data of printing production includes the following steps: A size position error-equipment and environment data mapping database is established based on the input and output of the decision tree model. A first color error-coating and environment data mapping database is established based on the input and output of the random forest regression model. A second color error-color-coated board and environment data mapping database is established based on the input and output of the random forest regression model. Obtaining the error tolerance thresholds stored in the database, including the size error tolerance threshold, the position error tolerance threshold, the first color error tolerance threshold, and the second color error tolerance threshold; Screening the combination of multi-dimensional impact data on printing production to be evaluated from the database of size and position error-equipment and environment data mapping, the first database of color error-coating and environment data mapping, and the second database of color error-color-coated board and environment data mapping; According to the combination of multi-dimensional impact data of each selected printing production to be evaluated, the corresponding comprehensive error value dYx to be evaluated is obtained; Determine the cost-effectiveness Xyb and implementation feasibility score Pf of each combination of multi-dimensional impact data of printing production to be evaluated; Determine each evaluation coefficient px; Among them, β1 is The weight factor of β2 is the weight factor of Xyb, and β3 is The weight factor, e is a natural constant; The combination of multi-dimensional impact data of printing production to be evaluated corresponding to the largest evaluation coefficient is used as the adjustment target, and the multi-dimensional impact data of printing production is optimized and regulated.

Images