Plastic substrate surface coating defect detection method based on LMC technology

Abstract

The invention relates to the technical field of visual detection of coating defects, in particular to a plastic base material surface coating defect detection method based on an LMC process, which comprises the following steps: acquiring a film layer morphology grayscale image of a plastic base material surface coating, constructing a pit direction difference degree of each pixel point, and obtaining a pit edge crack characteristic value of each pixel point; the method comprises the following steps of: obtaining a plurality of pixel points, further obtaining a pit characteristic value of each pixel point, extracting suspected pit characteristic points, obtaining pit dense saliency of each pixel point in combination with a position relationship between each pixel point and the suspected pit characteristic points, and obtaining a film pit saliency map by using a quaternary Fourier saliency detection algorithm. And carrying out image segmentation to obtain a coating pit defect area. According to the invention, the detection precision of coating defects can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of visual detection of coating defects, and in particular to a method for detecting coating defects on the surface of a plastic substrate based on the LMC process. Background Art

[0002] The LMC process utilizes atomic-level electromagnetic impact to deposit films on non-metallic surfaces. Coatings based on this process offer significant advantages in flatness, fully revealing surface details on plastic substrates while maintaining the original high gloss. However, the quality of LMC-based coatings on plastic substrates is significantly affected by the operating current, which can easily lead to defects such as pits, cracks, and holes in the film, affecting the uniformity and density of the film. Therefore, defect detection is necessary for coatings on plastic substrates.

[0003] Currently, most industrial visual inspection methods analyze the film topography characteristics of plastic substrate surfaces using film topography images to detect defects in plastic substrate surface coatings. However, during the LMC process, small pit defects may form due to the impact of metal atoms and metal ions. Since pit defects on plastic substrate surfaces are often relatively small, existing technologies are unable to accurately identify these pit defect characteristics in film topography images. This can lead to false detections and missed detections during defect detection after coating plastic substrate surfaces, affecting the accuracy of defect detection for plastic substrate surface coatings. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a method for detecting defects in coating on the surface of a plastic substrate based on the LMC process to solve the existing problems.

[0005] The present invention discloses a method for detecting defects in coatings on plastic substrate surfaces based on the LMC process, which employs the following technical solutions:

[0006] One embodiment of the present application provides a method for detecting coating defects on a plastic substrate surface based on an LMC process, comprising the following steps:

[0007] Obtaining grayscale images of the film layer morphology coated on the surface of the plastic substrate;

[0008] By analyzing the distribution and variation of grayscale texture in the local neighborhood of each pixel in the grayscale image of the film layer, the difference in the pit direction of each pixel is obtained.

[0009] According to the gradient change in the local neighborhood of each pixel point and the difference in the pit direction of each pixel point, the pit edge crack characteristic value of each pixel point is obtained;

[0010] Obtaining the pit feature value of each pixel point through the pit direction difference and the pit edge crack feature value, thereby extracting the suspected concave feature point of each pixel point, and obtaining the pit density significance of each pixel point based on the positional relationship between each pixel point and each suspected concave feature point;

[0011] The quaternion Fourier saliency detection algorithm is used to combine the grayscale value, gradient value, pit edge crack characteristic value and pit density saliency of each pixel to obtain the film layer pit saliency map, and the area of ​​coating pit defects is obtained through image segmentation.

[0012] Preferably, the method for obtaining the difference in the pit direction of each pixel point is:

[0013] G j =(Hg j -Hd j )×(Hg j +Hd j );

[0014] Where G j is the pit direction difference of the j-th pixel, Hg j and HD j are the horizontal texture complexity and vertical texture complexity of the j-th pixel area window respectively.

[0015] Preferably, with each pixel point as the center, a window of a preset size is used as the area window of each pixel point.

[0016] Preferably, the method for obtaining the horizontal texture complexity and the vertical texture complexity is:

[0017] Get each grayscale row vector and each grayscale column vector in the pixel area window, and take the mean of the permutation entropy of all grayscale row vectors in the area window of each pixel as the horizontal texture complexity of the area window of each pixel, and take the mean of the permutation entropy of all grayscale column vectors in the area window of each pixel as the vertical texture complexity of the area window of each pixel.

[0018] Preferably, the method for obtaining the pit edge crack characteristic value of each pixel point is:

[0019]

[0020] Where D j is the pit edge crack characteristic value of the j-th pixel point, n is the number of neighboring pixels in the j-th pixel area window, F j and G j are the gradient amplitude and pit direction difference of the j-th pixel, F j,g and G j,gare the gradient amplitude and pit direction difference of the g-th pixel point in the j-th pixel point area window, respectively; the gradient amplitude of the pixel point is obtained through the edge detection algorithm.

[0021] Preferably, the pit characteristic value of each pixel point is the sum of the pit direction difference and the pit edge crack characteristic value of each pixel point.

[0022] Preferably, extracting suspected concave feature points of each pixel further includes:

[0023] The mean of the pit feature values ​​of all pixels is used as the screening threshold, and the pixels whose pit feature values ​​are greater than the screening threshold are used as target pixels. With each pixel as the center, the K target pixels closest to each pixel are selected as the K suspected pit feature points of each pixel.

[0024] Preferably, the method for calculating the concave pit density significance of each pixel point is:

[0025]

[0026] Where V j is the concave density saliency of the j-th pixel, exp() is an exponential function with a natural constant as the base, ds j,k is the Euclidean distance between the jth pixel and its kth suspected concave feature point, DS j,k is the mean of the pit feature values ​​of the j-th pixel and its k-th suspected pit feature point, and K is the number of suspected pit feature points.

[0027] Preferably, the method for obtaining the film layer pit significance map is:

[0028] Perform edge detection on the grayscale image of the film layer morphology to obtain a gradient image of the film layer morphology;

[0029] The grayscale value of each pixel in the film layer morphology grayscale image is replaced by the pit edge crack feature value and the pit density significance, and the replaced images are recorded as the pit edge crack feature map and the pit density feature map respectively;

[0030] The film layer morphology grayscale image, film layer morphology gradient map, pit edge crack feature map and pit dense feature map are used as the four feature maps in the phase spectrum model PQRT of quaternion Fourier transform, and the film layer pit saliency map is obtained using the quaternion Fourier saliency detection algorithm.

[0031] Preferably, the method for obtaining the region of the coating pit defect is: using a watershed segmentation algorithm to segment the film pit saliency map to obtain the segmented film pit defect region.

[0032] This application has at least the following beneficial effects:

[0033] By analyzing the pit defect characteristics after LMC coating, this application more accurately measures the edge crack characteristics and pit density characteristics of the pit defects on the film layer morphology, making the pit defect characteristics on the plastic substrate surface more accurate, and avoiding the problems of false detection and missed detection in subsequent pit defect detection on the plastic substrate surface after coating;

[0034] This application combines the edge crack characteristics and pit density characteristics of the pit defects on the film morphology, and uses the quaternary Fourier saliency detection algorithm to extract the film pit saliency map, making the pit defects on the film morphology more obvious and improving the saliency of the pit defects on the film morphology. The image segmentation algorithm is then used to more accurately identify the pit defect area on the film surface, thereby improving the accuracy of defect detection for coatings on plastic substrate surfaces.

[0035] The present application solves the problem of poor significance of pit defect features on the surface of plastic substrates, enabling existing segmentation algorithms to more accurately identify the pit defect features on the surface of the film layer, thereby accurately segmenting the pit defect area on the surface of the film layer, thereby solving the problem of false detection and missed detection when performing defect detection on the surface of plastic substrates after coating in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0037] Figure 1 A flowchart of the steps of the method for detecting defects in coating on a plastic substrate surface based on the LMC process provided in this application;

[0038] Figure 2 Schematic diagram of the process of obtaining the film layer pit significance map provided in this application. DETAILED DESCRIPTION

[0039] To further illustrate the technical means and effectiveness of this application's implementation of the intended invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of the plastic substrate surface coating defect detection method based on the LMC process proposed in this application. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0040] Unless otherwise defined, terms such as "comprises," "comprising," or any other variants thereof are intended to encompass non-exclusive inclusion, such that a circuit structure, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such article or device. In the absence of further restrictions, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the article or device comprising the element. In addition, the term "and\or" as used herein includes any and all combinations of one or more related listed items. All technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application pertains.

[0041] The specific scheme of the method for detecting defects in coating on the surface of a plastic substrate based on the LMC process provided by the present application is described in detail below with reference to the accompanying drawings.

[0042] An embodiment of the present application provides a method for detecting defects in coatings on plastic substrate surfaces based on the LMC process. For details, please refer to Figure 1 , including the following steps:

[0043] Step 1: Obtain a grayscale image of the film layer morphology coated on the surface of the plastic substrate.

[0044] In this embodiment, a method for detecting defects on the surface of a film coated using the LMC process is implemented through industrial visual inspection. First, a plastic substrate coated using the LMC process is obtained, and an image sensor is used to obtain an image of the film layer morphology of the film coated on the plastic substrate surface through visual inspection. In this embodiment, image sensors are mainly divided into CCD image sensors and CMOS image sensors. CCD image sensors have the characteristics of high sensitivity and low noise, while CMOS image sensors have the characteristics of low power consumption and high integration. In this embodiment, the CCD image sensor is selected to collect images of the film layer morphology.

[0045] In order to avoid the interference caused by external noise and improve the accuracy of image acquisition by intelligent sensors in industrial visual inspection, the film layer morphology image is converted into grayscale, and the grayscale converted film layer morphology image is input into the filtering algorithm. The filtering algorithm can be bilateral filtering, Gaussian filtering, and mean filtering. This embodiment does not make specific limitations. In this embodiment, bilateral filtering is selected for denoising, and the denoised image is recorded as a film layer morphology grayscale image, where bilateral filtering is a well-known technology and the specific process is not repeated.

[0046] Step 2: By analyzing the grayscale texture distribution and change differences in the local neighborhood of each pixel point in the film layer morphology grayscale image, the pit direction difference of each pixel point is obtained.

[0047] In the LMC process, the instability of the operating current during atomic-level electromagnetic impact affects the density of the film layer, forming pits of varying sizes and shapes. However, the changes in atomic-level electromagnetic impact are relatively small, which can easily reduce the significance of the pit defects on the film layer detected by the smart sensor. Consequently, existing technologies are unable to accurately identify the pit defect characteristics on the film surface in the film layer topography image, leading to false detections and missed detections when performing defect detection on the surface of plastic substrates after coating. Therefore, in order to improve the accuracy of defect detection on the surface of plastic substrates, it is necessary to more accurately identify the pit defect characteristics in the grayscale image of the film layer topography collected by the smart sensor.

[0048] Furthermore, to accurately analyze the pit defects in the film topography identified by the smart sensor, a 9×9 regional window is set centered around each pixel in the grayscale image of the film topography. The implementer can choose the size of the regional window based on actual conditions. Furthermore, because pit defects on the film often vary in size and shape, the directional differences between the horizontal and vertical texture changes near the pit defect location in the grayscale image of the film topography are significant.

[0049] Therefore, the grayscale row vectors and grayscale column vectors in the regional window are obtained by measuring the grayscale values ​​of each pixel point in the film layer morphology grayscale image. The grayscale row vectors and grayscale column vectors can respectively reflect the horizontal grayscale texture changes and vertical grayscale texture changes near each pixel point in the film layer morphology grayscale image. If the difference in complexity changes between the horizontal grayscale texture and the vertical grayscale texture near a pixel point in the film layer morphology grayscale image is greater, and the complexity of the horizontal grayscale texture change and the vertical grayscale texture change is higher, the more it can reflect the directional difference characteristics of the pit defect position in the film layer morphology grayscale image.

[0050] Furthermore, the mean of the permutation entropy of all grayscale row vectors in the regional window of each pixel point is recorded as the horizontal texture complexity of the regional window of each pixel point, and the mean of the permutation entropy of all grayscale column vectors in the regional window of each pixel point is recorded as the vertical texture complexity of the regional window of each pixel point. The horizontal texture complexity and the vertical texture complexity respectively reflect the complexity of the horizontal texture and the vertical texture in the regional window. The calculation of the permutation entropy is a well-known technology and the specific process will not be repeated here.

[0051] Through the above analysis, the difference in the pit direction of each pixel in the grayscale image of the film layer is calculated:

[0052] G j =(Hg j -Hd j )×(Hg j +Hd j);

[0053] Where G j is the pit direction difference of the j-th pixel, Hg j and HD j are the horizontal texture complexity and vertical texture complexity of the j-th pixel area window respectively.

[0054] Since the pit defects formed during the atomic-level electromagnetic impact of the LMC process have the characteristics of different sizes and irregular shapes, the complex changes between the horizontal texture and the vertical texture of the pit defect position in the grayscale image of the film morphology identified by the intelligent sensor will show large differences. This can effectively enhance the pit defect characteristics in the grayscale image of the film morphology, which is conducive to more accurate identification of pit defects in the surface coating of plastic substrates.

[0055] Step 3: According to the gradient change in the local neighborhood of each pixel point and the difference in the pit direction of each pixel point, the pit edge crack feature value of each pixel point is obtained.

[0056] Generally speaking, the instability phenomenon in the atomic-level electromagnetic impact process of the LMC process can easily lead to cracks at the edge of the pits on the film morphology identified by the smart sensor, which will cause obvious chaotic changes in the difference in the pit direction at the edge of the pit.

[0057] Furthermore, the film layer morphology grayscale image is input into the edge detection algorithm, and the edge detection algorithm is used to obtain the gradient amplitude of all pixel points in the film layer morphology grayscale image, and the image obtained after edge detection is used as the film layer morphology gradient map. The edge detection algorithm can be a Sobel edge detection algorithm or a Canny edge detection algorithm, which is not specifically limited in this embodiment. In this embodiment, the Canny edge detection algorithm is selected to obtain the gradient amplitude of all pixel points in the film layer morphology grayscale image and the film layer morphology gradient map. The Canny edge detection algorithm is a well-known technology, and the specific process will not be repeated.

[0058] If the gradient change in the regional window of a certain pixel point is greater and the change in the pit direction difference is more inconsistent, to a certain extent, it can be reflected that the characteristics of the pit defect with edge cracks are greater. On the contrary, when the pixel point is not at the pit defect position, the gradient change in the regional window is small and the change in the pit direction difference is not obvious. At this time, there is no characteristic of the pit defect with edge cracks.

[0059] Through the above analysis, the pit edge crack characteristic value of each pixel point in the film layer grayscale image is calculated:

[0060]

[0061] Where D jis the pit edge crack characteristic value of the j-th pixel point, n is the number of neighboring pixels in the j-th pixel area window, F j and G j are the gradient amplitude and pit direction difference of the j-th pixel, F j,g and G j,g are the gradient amplitude and pit direction difference of the g-th pixel in the j-th pixel area window, respectively.

[0062] Because the atomic-level electromagnetic impact force of the LMC process is too strong, pit defects are formed, and the edges of the pit positions are often easily accompanied by the characteristic phenomenon of edge cracks. Therefore, the pit edge crack feature value reflects the edge crack characteristics of the pit defect in the grayscale image of the film morphology identified by the intelligent sensor. The larger the edge crack characteristics on the pit defect, the more clearly the coating defect characteristics based on the LMC process can be highlighted, thereby achieving more accurate identification of the pit defect characteristics on the film morphology, thereby avoiding the problems of false detection and missed detection when performing defect detection on the surface of the plastic substrate after coating.

[0063] Furthermore, in this embodiment, for the convenience of understanding and expression, the grayscale value of each pixel point in the grayscale image of the film layer morphology is replaced with the corresponding pit edge crack feature value, and all pixel points in the grayscale image of the film layer morphology are traversed for replacement, and the replaced feature map is recorded as the pit edge crack feature map of the film layer morphology.

[0064] Step 4: Obtain the pit feature value of each pixel point through the pit direction difference and the pit edge crack feature value, so as to extract the suspected concave feature points of each pixel point, and obtain the pit density significance of each pixel point by combining the positional relationship between each pixel point and each suspected concave feature point.

[0065] Generally, pits are formed due to the unstable working current of the LMC process. When pits are formed, multiple pits are often formed in a local area instead of just one pit, and the position distance between different pits in the local area is relatively close, which reflects the characteristic phenomenon of dense pits in the film layer morphology image recognized by the smart sensor.

[0066] At the same time, if the pit direction difference characteristics and pit edge crack characteristics at the pixel position in the film layer morphology grayscale image are more obvious, the characteristics of the pit defect on the film layer morphology can be more highlighted. Therefore, the sum of the pit direction difference and the pit edge crack characteristic value of each pixel point in the film layer morphology grayscale image is recorded as the pit characteristic value of each pixel point in the film layer morphology grayscale image. The pit characteristic value reflects the characteristics of the pit defect in the film layer morphology grayscale image identified by the intelligent sensor. The larger the pit characteristic value, the more likely the pixel point is to be at the pit position.

[0067] Furthermore, in order to analyze the characteristic phenomenon of dense pits in the film morphology image identified by the intelligent sensor, combined with the coating characteristics of the LMC process, the pit defect characteristics on the film surface can be more accurately identified, and the average of the pit characteristic values ​​of all pixels in the film morphology grayscale image is used as the screening threshold, and the pixels with pit characteristic values ​​greater than the screening threshold are used as target pixels. In this embodiment, with each pixel as the center, the K target pixels closest to the center pixel are selected and recorded as the K suspected concave feature points of each pixel in the film morphology grayscale image. In this embodiment, the value of K is 20, and the implementer can determine the value according to the actual situation. At the same time, the distance calculation between pixels is an existing technology and will not be repeated in this embodiment.

[0068] Through the above analysis, the pit density significance of each pixel in the grayscale image of the film layer is calculated:

[0069]

[0070] Where V j is the pit density saliency of the j-th pixel, exp() is an exponential function with a natural constant as the base, ds j,k is the Euclidean distance between the jth pixel and its kth suspected concave feature point, DS j,k is the mean of the pit feature values ​​of the j-th pixel and its k-th suspected pit feature point, and K is the number of suspected pit feature points.

[0071] The pit density significance reflects the characteristic phenomenon of dense pits in the film morphology identified by the intelligent sensor after LMC coating. The greater the pit density significance, the more it can highlight the dense characteristics of pit defects after LMC coating. Combined with the coating characteristics of the LMC process, it is conducive to more accurate identification of pit defect characteristics on the film surface and improve the accuracy of defect detection for plastic substrate surface coating.

[0072] Furthermore, in this embodiment, for ease of understanding and expression, the grayscale value of each pixel in the grayscale image of the film layer morphology is similarly replaced with the corresponding pit density significance, and all pixel points in the grayscale image of the film layer morphology are traversed for replacement, and the feature map after replacement is recorded as the pit density feature map of the film layer morphology.

[0073] Step 5: Use the quaternion Fourier saliency detection algorithm to combine the grayscale value, gradient value, pit edge crack feature value and pit density saliency of each pixel to obtain the film layer pit saliency map, and obtain the area of ​​the coating pit defect through image segmentation.

[0074] During the coating process on the surface of a plastic substrate using the LMC process, smaller pit defects are formed due to the impact of metal atoms and metal ions. However, the pit defect characteristics on the surface of the plastic substrate identified by the smart sensor are often relatively small.

[0075] Therefore, this embodiment uses the quaternion Fourier saliency detection algorithm, and combines the pit characteristics after LMC process coating to improve the significance of pits in the film layer morphology, and uses the film layer morphology grayscale image, film layer morphology gradient map, pit edge crack feature map and pit density feature map as the four feature maps in the phase spectrum model PQRT of the quaternion Fourier transform, and inputs them into the quaternion Fourier saliency detection algorithm. The quaternion Fourier saliency detection algorithm is used to obtain the film layer pit saliency map. It should be noted that the quaternion Fourier saliency detection algorithm is a well-known technology, and the specific process will not be repeated. Among them, the schematic diagram of the process of obtaining the film layer pit saliency map is as follows: Figure 2 shown.

[0076] Finally, an image segmentation algorithm is used to obtain the area of ​​coating defects on the surface of the plastic substrate. The image segmentation algorithm can specifically be a region growing segmentation algorithm, an Otsu threshold segmentation algorithm, or a watershed segmentation algorithm. In this embodiment, a watershed segmentation algorithm is used to perform image segmentation on the film layer pit saliency map to segment the regional image of the film layer pit defects, thereby realizing defect detection of the coating on the surface of the plastic substrate. Among them, the watershed segmentation algorithm is a well-known technology, and the specific process will not be repeated here.

[0077] It is understood that references to "one embodiment" or "some embodiments" in the present specification mean that one or more embodiments of the present application include a particular feature, structure, or characteristic described in conjunction with that embodiment. Thus, if "in one embodiment," "in some embodiments," "in other embodiments," or "in other embodiments" appear in different places in this specification, they do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0078] It should be noted that the above-mentioned sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above description is of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous. At the same time, the size of the sequence number of each step in the embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments in this specification.

[0079] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for detecting coating defects on a plastic substrate surface based on the LMC process, characterized in that: The following steps are involved: Obtaining grayscale images of the film layer morphology coated on the surface of the plastic substrate; By analyzing the distribution and variation of grayscale texture in the local neighborhood of each pixel in the grayscale image of the film layer, the difference in the pit direction of each pixel is obtained. According to the gradient change in the local neighborhood of each pixel point and the difference in the pit direction of each pixel point, the pit edge crack characteristic value of each pixel point is obtained; Obtaining the pit feature value of each pixel point through the pit direction difference and the pit edge crack feature value, thereby extracting the suspected concave feature point of each pixel point, and obtaining the pit density significance of each pixel point based on the positional relationship between each pixel point and each suspected concave feature point; The quaternion Fourier saliency detection algorithm is used to combine the grayscale value, gradient value, pit edge crack characteristic value and pit density saliency of each pixel to obtain the film layer pit saliency map, and the area of ​​coating pit defects is obtained through image segmentation.

2. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 1, wherein: The method for obtaining the pit direction difference of each pixel point is: G j =(Hg j -Hd j )×(Hg j +Hd j ); Where G j is the pit direction difference of the j-th pixel, Hg j and HD j are the horizontal texture complexity and vertical texture complexity of the j-th pixel area window respectively.

3. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 2, wherein: With each pixel as the center, a window of preset size is used as the area window of each pixel.

4. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 2, wherein: The method for obtaining the horizontal texture complexity and the vertical texture complexity is: Get each grayscale row vector and each grayscale column vector in the pixel area window, and take the mean of the permutation entropy of all grayscale row vectors in the area window of each pixel as the horizontal texture complexity of the area window of each pixel, and take the mean of the permutation entropy of all grayscale column vectors in the area window of each pixel as the vertical texture complexity of the area window of each pixel.

5. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 3, wherein: The method for obtaining the pit edge crack characteristic value of each pixel point is: Where D j is the pit edge crack characteristic value of the j-th pixel point, n is the number of neighboring pixels in the j-th pixel area window, F j and G j are the gradient amplitude and pit direction difference of the j-th pixel, F j,g and G j,g are the gradient amplitude and pit direction difference of the g-th pixel point in the j-th pixel point area window, respectively; the gradient amplitude of the pixel point is obtained through the edge detection algorithm.

6. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 1, wherein: The pit characteristic value of each pixel point is the sum of the pit direction difference and the pit edge crack characteristic value of each pixel point.

7. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 1, wherein: The extracting of suspected concave feature points of each pixel further includes: The mean of the pit feature values ​​of all pixels is used as the screening threshold, and the pixels whose pit feature values ​​are greater than the screening threshold are used as target pixels. With each pixel as the center, the K target pixels closest to each pixel are selected as the K suspected pit feature points of each pixel.

8. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 1, wherein: The calculation method of the pit density significance of each pixel point is: Where V j is the concave density saliency of the j-th pixel, exp() is an exponential function with a natural constant as the base, ds j,k is the Euclidean distance between the jth pixel and its kth suspected concave feature point, DS j,k is the mean of the pit feature values ​​of the j-th pixel and its k-th suspected pit feature point, and K is the number of suspected pit feature points.

9. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 1, wherein: The method for obtaining the film layer pit saliency map is as follows: Perform edge detection on the grayscale image of the film layer morphology to obtain a gradient image of the film layer morphology; The grayscale value of each pixel in the film layer morphology grayscale image is replaced by the pit edge crack feature value and the pit density significance, and the replaced images are recorded as the pit edge crack feature map and the pit density feature map respectively; The film layer morphology grayscale image, film layer morphology gradient map, pit edge crack feature map and pit dense feature map are used as the four feature maps in the phase spectrum model PQRT of quaternion Fourier transform, and the film layer pit saliency map is obtained using the quaternion Fourier saliency detection algorithm.

10. The method for detecting defects in coating on a plastic substrate surface based on the LMC process according to claim 1, wherein: The method for obtaining the region of the coating pit defect is: using a watershed segmentation algorithm to segment the film layer pit saliency map to obtain the segmented film layer pit defect region.

Images