A method and device for detecting flexible printing ghosting based on SFR

Abstract

The present invention relates to a method for detecting ghosting in flexible printing based on SFR detection, including the following steps: obtaining a first image of a target detection object; performing a preprocessing operation on the first image to obtain a second image; calculating the SFR of the second image based on the ISO12233 international standard; and determining whether there is ghosting in the printed product according to the SFR. Based on the ISO12233 standard, the present invention performs image processing on the pattern of the flexible printed product to obtain an MTF curve, and then detects whether there is ghosting in the printed product, ensuring that even subtle ghosting in flexible printing can be detected online. This method can not only quickly detect the quality of printed products, but also improve the accuracy of detecting printing ghosting, so as to improve the working efficiency of printed products. The present invention is applicable to but not limited to all flexible printed products with solid-color backgrounds.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and in particular to a method and device for detecting flexible printing ghosting based on SFR. Background Art

[0002] Flexographic printing is a printing method that uses a flexible printing plate and transfers ink through an anilox roll. It is a type of relief printing process, abbreviated as flexography. The graphic part of the flexible printing plate is raised. During printing, the anilox roll evenly coats a certain thickness of ink layer on the graphic part of the printing plate, and then under the action of the pressure of the impression cylinder, the ink layer on the graphic part is transferred to the surface of the substrate to form clear graphics and texts. In recent years, flexographic printing technology has developed rapidly globally with the advantage of being green and environmentally friendly, and its application fields have also been continuously expanded. In the current era background of the parallel development of informatization and intelligence, the application of computer technology has undoubtedly written a new chapter for the growth of the manufacturing industry and created a new development direction. Therefore, how to use the weapon of intelligence to enable the manufacturing industry to successfully move towards a steady and high-speed development path has become a key issue in the research of the development of the manufacturing industry.

[0003] In high-speed printing production, various printing defects may occur. Printing ghosting is one of the most common printing faults at present, manifested as blurred text and lines. When severe, double lines appear in the text and lines, while in pictures, dot deformation occurs, the color reproduction of the printed product is poor, the whole picture becomes blurred, the tone and level change, resulting in a decline in product quality. When observed with a magnifying glass, it can be clearly found that the edges of the dots are not clear, and the phenomenon of double dots or dot trailing appears. During printing, the reasons for ghosting are poor paper flatness and uneven distribution of the overall characteristics of the paper, excessive ink viscosity, wear of the printing press cylinder bearings, and increased clearance of the printing press cylinder gears. The appearance of ghosting will lead to a decrease in the clarity of the entire picture, distortion of color reproduction, and loss of tone and level, and it is impossible to achieve faithful reproduction relative to the original manuscript, seriously affecting the quality of printed products.

[0004] In the actual printing process, in order to effectively control the defective product rate in the printing production process, strict quality monitoring work must be carried out on the printing production process. The traditional manual sampling detection method depends on people's subjective evaluation, resulting in insufficient detection of printing ghosting, and low efficiency, high cost, strong subjectivity, and high error rate. The traditional visual defect detection methods are as follows.

[0005] CN111127417A A method for detecting printing defects of flexible packaging rolls based on SIFT feature matching and improved SSD algorithm. This method combines traditional visual defect detection methods and deep learning detection methods, uses SIFT feature matching and consistency detection of acquired images; constructs an improved SSD network model, trains the model, and uses the trained model. If there are cases of mislabeling or missed labeling in a small number of annotations, it enters the manual detection process; manually collects the mislabeled or missed labeled images, makes them into a training set after marking the defect positions, and then retrains the network; then sends the pictures into the retrained model. If the error information can be successfully labeled, the printing defects of the flexible packaging rolls are completed.

[0006] CN112070716A An intelligent recognition method for printing defects based on deep learning. Collect images of printing defect areas and corresponding template images, and mark the defect types to form a training set; construct a deep learning network model for extracting printing defect features and train it; based on the oversampling method, expand the sample features of the category with fewer samples; construct an SVM classifier and train the SVM with the feature data after sample expansion; for the printing defect images to be recognized, use the trained feature extraction network model and SVM classifier to perform defect recognition.

[0007] The above two methods at least have high requirements for the angle and illumination of images, low recognition accuracy, and deep learning-based printing defect detection requires a large number of training samples and is difficult to detect a kind of defect in the subtle double images of flexible printing. Summary of the Invention

[0008] The purpose of the present invention is to at least solve one of the deficiencies of the prior art, and provide a method and device for detecting flexible printing double images based on SFR.

[0009] To achieve the above purpose, the present invention adopts the following technical solutions:

[0010] Specifically, a method for detecting flexible printing double images based on SFR is proposed, including the following:

[0011] Obtain a first image of the target detection object;

[0012] Perform preprocessing operations on the first image to obtain a second image;

[0013] Calculate the SFR of the second image based on the ISO12233 international standard;

[0014] Judge whether there is double image in the printed product according to the SFR.

[0015] Further, specifically, the method for calculating the SFR curve of the second image includes the following:

[0016] Perform linearization processing on the second image to obtain second image data;

[0017] Calculate the centroid of each row of the second image data, that is, the image centroid, to obtain the edge position of each row of data;

[0018] Perform linear regression on the obtained centroid to obtain corresponding edge data, and the edge data includes slope and intercept;

[0019] Redefine the ROI of the first image data according to the edge data, and obtain its edge spread function ESF according to the distance from the first image data to the edge;

[0020] Perform 4-fold super-sampling on the ESF to obtain a 4-fold ESF;

[0021] Perform differential operation on the 4-fold ESF to obtain a line spread function LSF;

[0022] Reduce the influence of image edges and noise on the LSF through Hamming window processing;

[0023] Perform discrete Fourier transform on the LSF processed by the Hamming window to obtain an optical transfer function OTF;

[0024] Obtain the SFR from the real part of the OTF.

[0025] Furthermore, specifically, the preprocessing operation includes the following,

[0026] Perform edge extraction on the first image through the Canny edge detection algorithm;

[0027] Perform Hough transform line detection processing on the first image after edge extraction processing;

[0028] Perform ROI region extraction on the first image after Hough transform line detection processing;

[0029] Perform dark current removal processing on the first image after ROI region extraction through the BackgroundSubtractorMOG2 image segmentation algorithm to obtain a second image.

[0030] Furthermore, specifically, the condition for judging whether there is ghosting in the printed product according to the SFR is that when the MTF value at a spatial frequency of 10 lp / mm is greater than 0.9 and the MTF value at a spatial frequency of 30 lp / mm is greater than 0.7, it is determined that there is no ghosting in the printed product.

[0031] The present invention also proposes a flexible printing ghosting detection device based on SFR detection, which applies a flexible printing ghosting detection method as described in any one of the above, including the following:

[0032] An image acquisition module for acquiring a first image of a target detection object;

[0033] A preprocessing module for preprocessing the first image to obtain a second image;

[0034] An SFR calculation module for calculating the SFR of the second image based on the ISO12233 international standard;

[0035] A ghosting detection module for judging whether there is ghosting in a printed product according to the SFR.

[0036] The present invention also proposes a computer-readable storage medium, the computer-readable storage medium stores a computer program, characterized in that when the computer program is executed by a processor, the steps of the above method are implemented.

[0037] The beneficial effects of the present invention are:

[0038] The present invention proposes a method for detecting printing ghosting by measuring the modulation transfer function (MTF) based on the spatial frequency response (SFR). Based on the ISO12233 standard, the present invention performs image processing on the pattern of a flexible printed product to obtain an MTF curve, and then detects whether there is ghosting in the printed product, ensuring that even subtle ghosting in flexible printing can be detected online. This method can not only quickly detect the quality of printed products, but also improve the accuracy of detecting printing ghosting, so as to improve the working efficiency of printed products. Description of the Drawings

[0039] By elaborating on the embodiments shown in conjunction with the drawings, the above and other features of the present disclosure will become more obvious. The same reference numerals in the drawings of the present disclosure represent the same or similar elements. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:

[0040] Figure 1 Shown is a flowchart of a method for detecting flexible printing ghosting based on SFR according to the present invention;

[0041] Figure 2 Shown is a schematic diagram of an example of a flexible printing ghosting image to which the method for detecting flexible printing ghosting based on SFR according to the present invention is applied. Detailed Embodiments

[0042] The concept, specific structure, and technical effects of the present invention will be clearly and completely described below in conjunction with the embodiments and the accompanying drawings to fully understand the purpose, solution, and effects of the present invention. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The same reference numerals used throughout the drawings indicate the same or similar parts.

[0043] Referring to Figure 1 and Figure 2 , Embodiment 1, the present invention proposes a method for detecting flexographic ghosting based on SFR, including the following:

[0044] Obtain a first image of the target object to be detected;

[0045] Perform a preprocessing operation on the first image to obtain a second image;

[0046] Calculate the SFR of the second image based on the ISO12233 international standard;

[0047] Judge whether there is ghosting in the printed product according to the SFR.

[0048] As a preferred embodiment of the present invention, specifically, the method for calculating the SFR curve of the second image includes the following:

[0049] Perform linearization processing on the second image to obtain second image data;

[0050] Calculate the centroid of each row of the second image data, that is, the image center of gravity, to obtain the edge position of each row of data;

[0051] Perform linear regression on the obtained centroid to obtain the corresponding edge data, and the edge data includes slope and intercept;

[0052] Redefine the ROI of the first image data according to the edge data, and obtain its edge spread function ESF according to the distance from the first image data to the edge;

[0053] Perform 4-fold supersampling on the ESF to obtain 4-fold ESF;

[0054] Perform differential operation on the 4-fold ESF to obtain the line spread function LSF;

[0055] Reduce the influence of image edges and noise on the LSF through Hamming window processing;

[0056] Perform discrete Fourier transform on the LSF after Hamming window processing to obtain the optical transfer function OTF;

[0057] Obtain the SFR from the real part of the OTF.

[0058] As a preferred embodiment of the present invention, specifically, the preprocessing operation includes the following:

[0059] Performing edge extraction on the first image through the Canny edge detection algorithm; Edge detection is a basic tool in graphic image processing, computer vision, and machine vision, commonly used for feature extraction and feature detection, aiming to detect edges or discontinuous regions with significant changes in a digital image and capture regions with sharp changes in brightness. Using the images of flexible printed products collected by a CCD industrial camera for preprocessing and applying the edge detection algorithm to obtain the edge features of the printed product images can reduce the amount of image data, filter out unnecessary information, leave the important structure of the image, and greatly improve the calculation speed of subsequent line detection;

[0060] Compared with the Sobel edge detection method, the Canny edge detection method is not easily affected by noise and can detect real weak edges. The Canny algorithm uses two different thresholds to detect strong edges and weak edges respectively. Only when the weak edges are connected to the strong edges, the weak edges are included in the output image. Therefore, using the Canny algorithm can ensure that the tiny ghosting features of flexible printing are not lost.

[0061] Performing Hough transform line detection processing on the first image after edge extraction processing; Screening the lines according to the line length. Hough transform line detection is a typical, fast, and reliable method for detecting lines. It realizes the fitting of lines by transforming the image coordinate space to the parameter space. Applying the Hough transform line detection to the image edge features obtained in step (1) to get the fitted lines, and then screening the straight line edges of the printed matter document or pattern and the hypotenuses suitable for the SFR algorithm according to the given line length parameter.

[0062] Using the Hough transform line detection for the edge features extracted by the Canny algorithm. The Hough transform line detection algorithm is to convert the line detection problem in the image space into the point detection problem in the parameter space and complete the line detection task by finding the peak value in the parameter space. Specifically, if the pixels in an image form a straight line, then the curves corresponding to these pixel coordinate values (x, y) in the parameter space must intersect at a point. Therefore, only by transforming all pixel points (coordinate values) in the image into curves in the parameter space and detecting the intersection points of the curves in the parameter space can the straight line be determined. Screening the lines fitted by the Hough transform according to the length, and retaining the straight line edges of the printed matter document or pattern and the hypotenuses suitable for the SFR algorithm.

[0063] Extract the ROI region from the first image after Hough transform line detection; select the ROI (Region Of Interest) in the vertical direction. According to the ISO12233 international standard, this patent selects the hypotenuse ROI region containing the vertical direction to calculate the horizontal SFR, as shown in the attached Figure 2 The red rectangular area of L3 in the resolution test chart. In the international standard, the resolution test chart is used to measure the resolution of an electronic still image camera. In the image of a flexible printed product, the resolution of the ghosting part is reduced. Therefore, by calculating the SFR of the selected ROI region, it can be determined whether the printing is ghosted. This is one of the innovative points of this patent.

[0064] For defining the ROI, there are mainly the following requirements: for the vertical hypotenuse, the width of the ROI should be less than the height; in the ROI, only one hypotenuse is allowed and it should be roughly centered; at the bottom or top, the smaller part cannot be less than 5 pixels; at the center position, the distance from the boundary to the middle hypotenuse needs to be between 20 and 60 pixels; the height needs to be between 80 and 300; avoid stains in the ROI and ensure that the image boundary is basically continuous.

[0065] Perform dark current removal processing on the first image after ROI region extraction through the BackgroundSubtractorMOG2 image segmentation algorithm to obtain the second image. Dark current refers to the fact that when there is no incident light on the sensor, there is a certain signal output. Due to the imbalance of each pixel, the dark current between pixels is also inconsistent, resulting in current noise. Since this patent can be applied to all flexible prints with a solid color background and it is desired that the ROI region selected in step (iii) only has black and white hypotenuses, the background / foreground segmentation algorithm based on Gaussian mixture is applied to remove the noise influence of the dark current with the background color, which can maximize the restoration of the true MTF curve.

[0066] As a preferred embodiment of the present invention, specifically, the condition for determining whether there is ghosting in the printed product according to the SFR is that when the MTF value at a spatial frequency of 10 lp / mm is greater than 0.9 and the MTF value at a spatial frequency of 30 lp / mm is greater than 0.7, it is determined that there is no ghosting in the printed product. Since the boundaries and contours in the ghosting part of the flexible printed image are relatively blurred, the calculated MTF curve decays rapidly at low frequencies. Therefore, it can be determined whether there is ghosting in the image data of the flexible printed product according to the low-frequency MTF value of the curve. Through experimental analysis, when the MTF value at a spatial frequency of 10 lp / mm is greater than 0.9 and the MTF value at a spatial frequency of 30 lp / mm is greater than 0.7, there is almost no ghosting in the printed product. Therefore, using this as the threshold to determine the condition for no ghosting gives a better judgment effect.

[0067] The present invention also provides a flexible printing ghosting detection device based on SFR detection, which applies a flexible printing ghosting detection method as described in any one of the above, including the following:

[0068] An image acquisition module for acquiring a first image of a target detection object;

[0069] A preprocessing module for performing preprocessing operations on the first image to obtain a second image;

[0070] An SFR calculation module for calculating the SFR of the second image based on the ISO12233 international standard;

[0071] A ghosting detection module for determining whether there is ghosting in a printed product according to the SFR.

[0072] The present invention also provides a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the steps of the above method are implemented.

[0073] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in this embodiment.

[0074] In addition, in each embodiment of the present invention, the functional modules may be integrated into one processing module, or each module may exist physically alone, or two or more modules may be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0075] When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or system, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0076] Although the description of the present invention has been quite detailed and several of the described embodiments have been described, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but rather should be regarded as providing a broad interpretation of these claims in light of the prior art by reference to the appended claims, thereby effectively covering the intended scope of the present invention. In addition, the present invention has been described above in terms of embodiments foreseeable by the inventor for the purpose of providing a useful description, and non-substantive modifications to the present invention that are not currently foreseeable may still represent equivalent modifications of the present invention.

[0077] As described above, these are only the preferred embodiments of the present invention. The present invention is not limited to the above-described embodiments. As long as it achieves the technical effects of the present invention by the same means, it should fall within the protection scope of the present invention. Within the protection scope of the present invention, various different modifications and variations can be made to its technical solutions and / or embodiments.

Claims

1. A method for detecting ghosting in flexible printing based on SFR, characterized in that Including the following: Obtain a first image of the target detection object; Perform a preprocessing operation on the first image to obtain a second image; Calculate the SFR of the second image based on the ISO12233 international standard; Judge whether there is ghosting in the printed product according to the SFR; Specifically, the method for calculating the SFR curve of the second image includes the following: Perform linearization processing on the second image to obtain second image data; Calculate the centroid of each row of the second image data, that is, the image centroid, to obtain the edge position of each row of data; Perform linear regression on the obtained centroid to obtain corresponding edge data, and the edge data includes slope and intercept; Relocate the ROI of the first image data according to the edge data, and obtain its edge spread function ESF according to the distance from the first image data to the edge; Perform 4-fold supersampling on the ESF to obtain a 4-fold ESF; Perform a difference operation on the 4-fold ESF to obtain a line spread function LSF; Reduce the influence of image edges and noise on the LSF through Hamming window processing; Perform a discrete Fourier transform on the LSF after Hamming window processing to obtain an optical transfer function OTF; Obtain the SFR from the real part of the OTF; Specifically, the preprocessing operation includes the following: Perform edge extraction on the first image through the Canny edge detection algorithm; Perform Hough transform line detection processing on the first image after edge extraction processing; Perform ROI region extraction on the first image after Hough transform line detection processing; Perform dark current removal processing on the first image after ROI region extraction through the BackgroundSubtractorMOG2 image segmentation algorithm to obtain a second image; Specifically, the condition for judging whether there is ghosting in the printed product according to the SFR is that when the MTF value at a spatial frequency of 10 lp / mm is greater than 0.9 and the MTF value at a spatial frequency of 30 lp / mm is greater than 0.7, it is determined that there is no ghosting in the printed product.

2. A flexible printed ghosting detection device based on SFR detection, characterized in that, Apply a method for detecting flexible printing ghosting based on SFR as described in claim 1 above, including the following: An image acquisition module for obtaining a first image of the target detection object; A preprocessing module for performing a preprocessing operation on the first image to obtain a second image; An SFR calculation module for calculating the SFR of the second image based on the ISO12233 international standard; A ghosting detection module for judging whether there is ghosting in the printed product according to the SFR.

3. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in claim 1.

Images