A method and apparatus for detecting defects in a road surface

Abstract

The application provides a glue path defect detection method and device, the method comprising: acquiring a gray image containing a glue path to be detected; determining the gray value of each first pixel point in a target line in the gray image, the target line penetrating the glue path along the glue path direction of the glue path; determining the edge pixel point in the first pixel point according to the gray value of each first pixel point, the difference between the gray value of the edge pixel point and the gray value of the corresponding reference pixel point being greater than a first threshold, the corresponding reference pixel point of the edge pixel point and the edge pixel point being located in the same target line, and the edge pixel point being adjacent to the corresponding reference pixel point; and determining whether the glue path has a glue path defect based on the edge pixel point. The embodiments can comprehensively detect each pixel point of the glue path through the target line, can determine the edge pixel point if the glue path has a region where the glue is not coherent, can determine whether the glue path has a glue path defect according to the edge pixel point, and can detect each type of glue path defect existing in the glue path.

Description

Technical Field

[0001] This application relates to the field of adhesive circuit calibration technology, and in particular to a method and apparatus for detecting adhesive circuit defects. Background Technology

[0002] In the industrial production process of products, dispensing technology is widely used. Dispensing, also known as gluing, coating, potting, or dripping, involves applying, potting, or dripping glue or other liquids onto a product to achieve functions such as adhesion, potting, insulation, fixation, and surface smoothing.

[0003] After applying adhesive to a product, the adhesive area on the product (called the adhesive path, for example) Figure 1 The black area in the image may contain defects. These defects include straight-edge breakage defects and non-straight-edge breakage defects. For example... Figure 1 As shown, a glue break region with two straight edges perpendicular to the glue path direction (i.e., a non-glue region) is called a straight-edge glue break. Other types of glue break regions are called non-straight-edge glue breaks. Non-straight-edge glue breaks include beveled glue breaks and curved glue breaks. Among them, a glue break region with two straight edges forming a single angle with the glue path direction is called a beveled glue break, and a glue break region with two curved edges forming multiple angles with the glue path direction is called a curved glue break.

[0004] Since defects in the adhesive path can affect product quality, there is a need for a solution that can detect these defects. Summary of the Invention

[0005] This application provides a method for detecting defects in adhesive circuits, which can detect various types of defects in adhesive circuits and improve the accuracy of adhesive circuit defect detection.

[0006] Firstly, a method for detecting defects in adhesive circuits is provided, including: Obtain a grayscale image containing the adhesive path to be detected; Determine the grayscale value of each first pixel point within the target line in the grayscale image; the target line penetrates the adhesive path along the adhesive path direction. Based on the gray values ​​of each first pixel, edge pixels in the first pixel are determined. The difference between the gray value of each edge pixel and the gray value of the corresponding reference pixel is greater than the first threshold. The reference pixel corresponding to the edge pixel is located on the same target line as the edge pixel, and the edge pixel is adjacent to the corresponding reference pixel. Based on edge pixels, determine whether there are adhesive defects in the adhesive path.

[0007] In a feasible design, determining whether adhesive path defects exist based on edge pixels includes: Determine each second pixel in each first pixel point, where the second pixel point is a non-glue pixel point located on the line segment formed by two edge pixels; Select two edge pixels and one pixel from each of the second pixels as the first target pixel; Based on the first target pixel, and according to the first preset search length, the search is performed along the first direction perpendicular to the adhesive path and the second direction perpendicular to the adhesive path, respectively. The searched pixels are the third pixels corresponding to the first target pixel. Based on the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel, determine whether the glue path defect type includes straight-edge glue breakage.

[0008] In a feasible design, determining whether adhesive path defects exist based on edge pixels includes: Determine the center pixel of the line segment formed by two edge pixels; Determine each fourth pixel in each first pixel, where the fourth pixel is located on the line segment formed by the center pixel and the two edge pixels respectively; If each fourth pixel point includes glue pixels, the glue path defect type is determined to exclude the beveled glue breakage. If each fourth pixel is a non-glue pixel, determine two starting pixels located next to the center pixel along the first direction perpendicular to the glue path and the second direction perpendicular to the glue path, respectively. If both starting pixels are non-glue pixels, then starting from the two starting pixels, search for the second target pixel corresponding to the starting pixel along two directions parallel to the glue path. The second target pixel is a non-glue pixel, and the distance between the second target pixel and the starting point is within the second preset search length. After finding glue pixels within the second preset search length, a new center pixel is determined. The new center pixel is the center pixel of each second target pixel, and a new starting pixel corresponding to the new center pixel is determined. If all new starting pixels are non-glue pixels, starting from the new starting pixels, search for the second target pixels corresponding to the new starting pixels again along two directions parallel to the glue path, until no glue pixels are found within the second preset search length of the starting point. If all starting pixels and all second target pixels are non-adhesive pixels, the glue path defect type is determined to include beveled glue breakage. If each starting pixel and each second target pixel includes glue pixels, the glue path defect type is determined to exclude beveled glue breaks.

[0009] In a feasible design, determining whether adhesive path defects exist based on edge pixels includes: Determine the fifth pixel on each line segment formed by two edge pixels; If each fifth pixel point includes glue pixels, the glue path defect type is determined to exclude curved glue breakage. If each fifth pixel is a non-glue pixel, determine the third target pixel in each fifth pixel. The third target pixel is located on a straight line perpendicular to the glue path direction. Based on the preset adhesive path width, obtain two fourth target pixels that are located on the straight line and outside the edge of the adhesive path; Using a path search algorithm, a path search is performed starting from the third target pixel and ending at the two fourth target pixels. If the path search is successful, the type of adhesive defect in the adhesive path is determined to include curved adhesive breakage; If the path search fails, the type of adhesive defect in the determined adhesive path does not include curved adhesive breaks.

[0010] In a feasible design, if the glue path defect type includes straight-edge glue breakage, the method also includes: The pixel width of the straight-cut adhesive is determined based on the pixel coordinates of the two edge pixels.

[0011] In a feasible design, if the glue path defect type includes straight-edge glue breakage, the method also includes: The boundary for connected component analysis is determined based on the pixel width, the preset pixel height of the adhesive path, and the pixel coordinates of the two edge pixels. The area of ​​the straight-cut glue break within the boundary is determined based on connected component analysis.

[0012] In a feasible design, it also includes: After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the first end that was previously searched in the third direction parallel to the adhesive path is compared with the second target pixel point located at the first end that was searched in this search, and the second target pixel point located at the first end is re-determined based on the comparison result. After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the second end that was previously searched in the fourth direction parallel to the adhesive path is compared with the second target pixel point located at the second end that was searched in this search. The second target pixel point located at the second end is re-determined based on the comparison result. The first direction and the second direction are opposite directions.

[0013] In a feasible design, after determining that the glue path defect types include beveled glue breaks, the method further includes: The cumulative sum of the number of each second target pixel found in each search process is obtained, and the cumulative sum is determined as the area of ​​the angled glue break. Alternatively, the number of center pixels found can be determined as the height of the beveled glue cut. Alternatively, the circumscribed rectangle of the angled glue cut can be determined based on the second target pixel at the first end and the second target pixel at the second end.

[0014] In a feasible design, if the adhesive path defect type includes curved adhesive breaks, the method also includes: Two search paths are determined, one starting from the third target pixel and the other ending at the two fourth target pixels. Take the third target pixel as the first endpoint, and take the pixels adjacent to the first endpoint on each search path as the second endpoint, and determine the normals of the line segments formed by the first endpoint and each second endpoint. Starting from the first endpoint, search for non-glue pixels along both directions of each normal. If glue pixels are found along both directions of the normal, determine the two non-glue pixels located at the curved glue break edge that are adjacent to the glue pixels found. The second endpoint is used as the new first endpoint, and the pixel adjacent to the new first endpoint is used as the new second endpoint. The search is performed again to determine the two new non-glue pixels located at the edge of the curved glue break, until the glue pixel is not found in either direction of the normal within the third preset search length starting from the first endpoint. Based on the searched non-glue pixels located at the edge of the curved glue break, determine the two target non-glue pixels that are furthest apart; Determine the width and circumscribed rectangle of the curved glue break based on two target non-glue pixels.

[0015] Secondly, a device for detecting defects in adhesive circuits is provided, comprising: The image acquisition module is used to acquire and send a grayscale image containing the adhesive path to be detected to the grayscale value determination module; The grayscale value determination module is used to determine and send the grayscale value of each first pixel in the target line in the grayscale image to the pixel determination module. The target line runs through the adhesive path along the adhesive path direction. The pixel determination module is used to determine and send the edge pixels in the first pixel to the glue path detection module based on the gray value of each first pixel. The difference between the gray value of the edge pixel and the gray value of the corresponding reference pixel is greater than a first threshold. The reference pixel corresponding to the edge pixel is located on the same target line as the edge pixel, and the edge pixel is adjacent to the corresponding reference pixel. The adhesive path detection module is used to determine whether there are adhesive path defects based on edge pixels.

[0016] In the above embodiments of this application, after acquiring a grayscale image containing the adhesive path to be detected, the grayscale value of each first pixel within the target path in the grayscale image is determined. To comprehensively detect each pixel of the adhesive path, the target path needs to penetrate the adhesive path along its direction. Based on the grayscale values ​​of each first pixel, if edge pixels among the first pixels can be identified, it means that there are areas of discontinuous adhesive in the adhesive path. Therefore, the presence of adhesive path defects can be further determined based on the edge pixels, thus achieving the detection of the adhesive path.

[0017] Furthermore, edge pixels are pixels on the edges of areas without adhesive in the adhesive path. These areas may be caused by different types of adhesive path defects. The solution provided in this application can detect various types of adhesive path defects by detecting edge pixels. Attached Figure Description

[0018] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of an adhesive path with defects provided in an exemplary embodiment of this application; Figure 2 This is a flowchart of an exemplary embodiment of the glue path defect detection method provided in this application; Figure 3 This is a schematic diagram of a straight-edge glue breakage detection provided by an exemplary embodiment of this application; Figure 4 This is a schematic diagram illustrating the effect of adhesive circuit defect detection provided in an exemplary embodiment of this application; Figure 5 This is a schematic flowchart illustrating a beveled glue breakage detection method provided in an exemplary embodiment of this application; Figure 6 This is a schematic diagram of a beveled glue breakage detection provided in an exemplary embodiment of this application; Figure 7 This is a schematic flowchart illustrating a curved adhesive breakage detection method provided in an exemplary embodiment of this application; Figure 8 This is a schematic diagram of a curved adhesive breakage detection provided in an exemplary embodiment of this application; Figure 9 This is a schematic diagram illustrating defect information detection for curved adhesive breaks provided in an exemplary embodiment of this application; Figure 10This is a flowchart of another example of a glue path defect detection method provided in an exemplary embodiment of this application; Figure 11 This is a schematic diagram of an exemplary adhesive circuit defect detection device provided in an exemplary embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] Figure 2 This is a flowchart of an exemplary embodiment of the glue path defect detection method provided in this application, as follows: Figure 2 As shown, this application provides a method for detecting defects in adhesive circuits, which can be applied to electronic devices (such as cameras or host computers communicating with cameras). The method includes the following steps: S110, acquire a grayscale image containing the adhesive path to be detected.

[0022] Typically, an image containing the adhesive path to be detected can be obtained by means of photography, and then the image can be converted into a corresponding grayscale image.

[0023] For example, an image containing the adhesive path to be detected is converted into a grayscale image in the following manner: Obtain an initial image containing the adhesive path to be detected; The initial image is processed using an image segmentation method to obtain a grayscale image containing the adhesive path to be detected.

[0024] Image segmentation methods include, but are not limited to, color extraction. For example... Figure 1 As shown, in the grayscale image, the pixel value of glue pixels is 0, and the pixel value of non-glue pixels is 255. It should be understood that the pixel value of glue pixels in the processed grayscale image can also be 255, and the pixel value of non-glue pixels can be 0. That is, this application does not limit the pixel values ​​of glue pixels and non-glue pixels, as long as they serve a distinguishing function.

[0025] In the above example, the image containing the glue path to be detected is converted into a grayscale image through image segmentation. This ensures that the grayscale image contains only glue pixels and non-glue pixels, which facilitates subsequent processing of each pixel.

[0026] S120, determine the grayscale value of each first pixel point within the target line in the grayscale image.

[0027] The target route runs through the rubber road along the rubber road direction.

[0028] In one feasible design, if the adhesive path to be detected is a straight line, then the target path is also a straight line and parallel to the direction of the adhesive path. If the adhesive path to be detected has other shapes, then the target path is also a shape adapted to the adhesive path to be detected; for example, if the adhesive path to be detected is curved, then the target path is also curved.

[0029] In addition, there can be one or more target lines. Generally, the higher the required detection accuracy, the more target lines are needed. If there is only one target line, it is usually located in the middle of the adhesive path.

[0030] For example, when a line-shaped control module is provided in the camera and the adhesive path defect detection method is applied to a host computer, the line-shaped control module is used to define the position of the target line. The host computer obtains the position of the target line in the following way to determine the grayscale value of each first pixel point within the target line in the grayscale image: The camera acquires an initial image containing the adhesive path to be detected; The camera uses a line-shaped control to determine the position of the target line in the initial image; The camera sends the initial image and the corresponding location of the target line to the host computer. The host computer receives the initial image and the corresponding location of the target line.

[0031] S130, determine the edge pixels in the first pixel based on the gray value of each first pixel.

[0032] Specifically, the grayscale value of each edge pixel must differ from the grayscale value of its corresponding reference pixel by more than a first threshold. The reference pixel and the edge pixel must be located on the same target line, and the edge pixel must be adjacent to its corresponding reference pixel. The first threshold is used to distinguish between the reference pixel and its adjacent edge pixels and can be set as needed.

[0033] It should be understood that edge pixels are the pixels at the edges of areas with no adhesive in the adhesive path. If no such edge pixels are detected, it indicates that the current adhesive path is free of defects. Additionally, straight-edged, angled, and curved adhesive breaks are typically defects that penetrate the entire adhesive path; see, for example... Figure 1 Straight-edge breakage, angled breakage, and curved breakage are all defects that penetrate the adhesive path in the direction perpendicular to the adhesive path. Therefore, in the solution provided in this application embodiment, if the area without adhesive in the adhesive path is small, it may not be considered a defect of straight-edge breakage, angled breakage, or curved breakage. In this case, after determining the edge pixels, it is necessary to further detect the adhesive path according to subsequent steps.

[0034] For example Figure 1 As shown, taking a first threshold of 254 as an example, let's explain the two edge pixels of a straight-edge glue break. If the reference pixel is a non-glue pixel on the target line (the pixel value of the reference pixel is 255), then the pixel value of the corresponding edge pixel is 0; if the reference pixel is a glue pixel on the target line (the pixel value of the reference pixel is 0), then the pixel value of the corresponding edge pixel is 255. Therefore, the pixel values ​​of the two edge pixels have the following four possibilities: (1) The reference pixel on the first side is 0, the edge pixel on the first side is 255, the reference pixel on the second side is 0, and the edge pixel on the second side is 255; (2) The reference pixel on the first side is 255, the edge pixel on the first side is 0, the reference pixel on the second side is 0, and the edge pixel on the second side is 255; (3) The reference pixel on the first side is 0, the edge pixel on the first side is 255, the reference pixel on the second side is 255, and the edge pixel on the second side is 0; (4) The reference pixel on the first side is 255, the edge pixel on the first side is 0, the reference pixel on the second side is 255, and the edge pixel on the second side is 0.

[0035] Two edge pixels can be determined based on any of the above situations, and this application does not limit this.

[0036] S140, based on the edge pixels, determine whether there is a glue path defect in the glue path.

[0037] In a feasible design, this application provides the following method for detecting straight-edge glue breakage: Determine each second pixel in each first pixel point, where the second pixel point is a non-glue pixel point located on the line segment formed by two edge pixels; Select two edge pixels and one pixel from each of the second pixels as the first target pixel; Based on the first target pixel, and according to the first preset search length, the search is performed along the first direction perpendicular to the adhesive path and the second direction perpendicular to the adhesive path, respectively. The searched pixels are the third pixels corresponding to the first target pixel. Based on the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel, determine whether the glue path defect type includes straight-edge glue breakage.

[0038] The first preset search length is used to specify the maximum number of pixels to be searched along the first and second directions, so as to avoid the search range exceeding the adhesive path.

[0039] It should be understood that the first direction and the second direction are opposite, for example, see Figure 3 If the first direction is perpendicular to the glue path direction and is upward, then the second direction is perpendicular to the glue path direction and is downward.

[0040] For example, determining whether the glue path defect type includes straight-edge glue breakage can be achieved in the following way: Obtain the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel; If the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel is 0, the glue path defect type is determined to include straight-edge glue breakage. If the difference between the grayscale value of the first target pixel and the grayscale value of the corresponding third pixel is not zero, the glue path defect type is determined to be excluding straight-edge glue breakage. In this example, if the difference between the grayscale value of the first target pixel and the grayscale value of the corresponding third pixel is not zero, it indicates that glue pixels have been found.

[0041] For example, during the search for a third pixel along the first and second directions for the first target pixel, the difference between the grayscale value of the first target pixel and the grayscale value of the third pixel is calculated for each third pixel found. If the difference is 0, the subsequent steps are performed; if the difference is not 0, it is determined that the glue path defect type does not include straight-edge glue breakage, and the straight-edge glue breakage detection ends. Using this method, once it is determined that the glue path defect type does not include straight-edge glue breakage, the subsequent straight-edge glue breakage detection step is unnecessary, saving time.

[0042] The above example can be combined with the search process for the third pixel. Figure 3 To facilitate understanding, and to distinguish between the target circuit and the adhesive circuit, the adhesive circuit is represented by a gray area. For example... Figure 3 As shown, first, two edge pixels are determined, namely edge pixel a and edge pixel b in the diagram. Then, the first target pixel located between edge pixel a and edge pixel b on the target path is determined. Finally, following the direction from edge pixel a to edge pixel b, the third pixel corresponding to the first target pixel is searched, and the difference between the grayscale value of the first target pixel and the grayscale values ​​of the corresponding third pixels is calculated. Figure 3 As shown, since the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel is 0, the glue path defect type is determined to include straight-edge glue breakage.

[0043] The above example, after determining each first target pixel between two edge pixels, searches for each third pixel along the direction perpendicular to the adhesive path for each first target pixel. Based on the difference between the gray values ​​of each third pixel and the gray values ​​of the first target pixels, it can accurately determine whether the adhesive path defect type includes straight-edge adhesive breakage.

[0044] In a feasible design, the pixel width of the straight-edge adhesive break is determined as follows: The pixel width of the straight-cut adhesive is determined based on the pixel coordinates of the two edge pixels.

[0045] For example, the absolute value of the difference between the pixel coordinates of two edge pixels is used as the pixel width of the straight edge of the glue.

[0046] After detecting a straight-edge glue break defect, it is usually necessary to mark the circumscribed rectangle of the glue break in the image. The circumscribed rectangle can be determined based on the location of the target circuit, the location of the glue path, and the pixel width of the glue break.

[0047] For example, the target line is the center line of the adhesive path in the horizontal direction. Two edge pixels are used as the center points of the two parallel sides of the circumscribed rectangle. The preset adhesive path width is used as the width of the circumscribed rectangle, and the pixel distance between the two edge pixels is used as the length of the circumscribed rectangle. The circumscribed rectangle for the straight-cut adhesive break is determined as follows: Figure 4 As shown.

[0048] Furthermore, the area of ​​the straight-edge break is determined by the following method: The boundary for connected component analysis is determined based on the pixel width, the preset pixel height of the adhesive path, and the pixel coordinates of the two edge pixels. The area of ​​the straight-cut glue break within the boundary is determined based on connected component analysis.

[0049] The above example demonstrates how connected component analysis can quickly and accurately determine the area of ​​a straight section of adhesive, and it is applicable to adhesive paths of various shapes.

[0050] In a feasible design, this application provides the following method for detecting glue breakage at the bevel, such as... Figure 5 As shown: Determine the center pixel of the line segment formed by two edge pixels; Determine each fourth pixel in each first pixel, where the fourth pixel is located on the line segment formed by the center pixel and the two edge pixels respectively; If each fourth pixel point includes glue pixels, the glue path defect type is determined to exclude the beveled glue breakage. If each fourth pixel is a non-glue pixel, determine two starting pixels located next to the center pixel along the first direction perpendicular to the glue path and the second direction perpendicular to the glue path, respectively. If both starting pixels are non-glue pixels, then starting from the two starting pixels, search for the second target pixel corresponding to the starting pixel along two directions parallel to the glue path. The second target pixel is a non-glue pixel, and the distance between the second target pixel and the starting point is within the second preset search length. After finding glue pixels within the second preset search length, a new center pixel is determined. The new center pixel is the center pixel of each second target pixel, and a new starting pixel corresponding to the new center pixel is determined. If all new starting pixels are non-glue pixels, starting from the new starting pixels, search for the second target pixels corresponding to the new starting pixels again along two directions parallel to the glue path, until no glue pixels are found within the second preset search length of the starting point. If all starting pixels and all second target pixels are non-adhesive pixels, the glue path defect type is determined to include beveled glue breakage. If each starting pixel and each second target pixel includes glue pixels, the glue path defect type is determined to exclude beveled glue breaks.

[0051] The second preset search length is used to specify the maximum number of pixels to be searched in two directions parallel to the adhesive path, and can be set according to actual needs.

[0052] The next position of the center pixel refers to the position of the next pixel in the first and second directions.

[0053] It should be noted that, Figure 5 Taking the first direction perpendicular to the glue path as an example, the detection process for beveled glue breakage is explained. For the detection process in the second direction perpendicular to the glue path, please refer to [link to relevant documentation]. Figure 5 That's all.

[0054] The following is combined with Figure 6 The detection process for beveled glue breakage is illustrated by example. Figure 6 For explanations regarding each direction, please refer to [link / reference]. Figure 3 This will not be elaborated upon here. Figure 6As shown, first, determine the center pixel e1 of the line segment formed by the two edge pixels. Then, determine the fourth pixel among the first pixels. Since each fourth pixel is a non-glue pixel, the subsequent steps are explained using the first direction perpendicular to the glue path as an example. Determine the starting pixel e2 located next to the center pixel e1. Since the starting pixel e2 is a non-glue pixel, starting from the starting pixel e2, search for the second target pixel corresponding to the starting pixel e2 along two directions parallel to the glue path. After finding glue pixels c1 and c2 within the second preset search length, determine a new center pixel e3. The new center pixel e3 is the center pixel of each of the second target pixels, and determine the new starting pixel of the new center pixel. Repeat the above search process until the two edges of the glue path (i.e., glue path edge 1 and glue path edge 2) are found. Figure 6 As shown, since each starting pixel and each second target pixel are non-adhesive pixels, the glue path defect type is determined to include beveled glue breakage.

[0055] The example above starts with the center pixels located between the two edges of the adhesive path and searches for the second target pixels in two directions parallel to the adhesive path until the glue pixel is found. Finally, it determines whether each starting pixel and each second target pixel is a non-glue pixel, which can accurately determine whether the adhesive path defect type includes beveled glue breakage.

[0056] After detecting a defect of beveled glue breakage in the glue path, in order to facilitate the determination of relevant information about the beveled glue breakage and to facilitate subsequent research by the inspectors, it is necessary to obtain the second target pixel points at both ends of the beveled glue breakage.

[0057] In a feasible design, the second target pixel points located at both ends of the beveled glue cut are determined by the following method: After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the first end that was previously searched in the third direction parallel to the adhesive path is compared with the second target pixel point located at the first end that was searched in this search, and the second target pixel point located at the first end is re-determined based on the comparison result. After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the second end that was previously searched in the fourth direction parallel to the adhesive path is compared with the second target pixel point located at the second end that was searched in this search. The second target pixel point located at the second end is re-determined based on the comparison result. The first direction and the second direction are opposite directions.

[0058] It should be noted that if all fourth pixels are non-glue pixels, then the second target pixels located at both ends initially determined are the pixels located at both ends among the searched fourth pixels.

[0059] In the example above, after each search along two directions parallel to the adhesive path, the second target pixels at both ends are updated in real time by comparing them with the second target pixels at both ends found in the current search. This continues until the final search, at which point the determined second target pixels at both ends are identified as the second target pixels at the ends of the beveled adhesive break. This allows for the simultaneous detection of a beveled adhesive break defect and the identification of the second target pixels at both ends of the beveled adhesive break, facilitating immediate calculation of the circumscribed rectangle of the beveled adhesive break.

[0060] In a feasible design, the area of ​​the beveled cut is determined as follows: The sum of the number of each second target pixel found in each search process is obtained, and the sum is determined as the area of ​​the beveled glue break.

[0061] In a feasible design, the height of the beveled cut is determined as follows: The number of center pixels found is determined as the height of the beveled glue cut.

[0062] In one feasible design, after determining the second target pixels at both ends of the beveled adhesive cut, the circumscribed rectangle of the beveled adhesive cut is determined as follows: The second target pixel at the first end and the second target pixel at the second end are taken as the two diagonally opposite vertices of the outer rectangle of the beveled glue cut. The width of the preset glue path is used as the width of the outer rectangle of the beveled glue cut, thus determining the outer rectangle of the beveled glue cut (e.g., ...). Figure 4 (As shown).

[0063] The above method utilizes the pixels obtained during the search process to accurately and quickly calculate the defect information after determining that there is a beveled glue breakage defect in the glue path.

[0064] In a feasible design, this application provides the following method for detecting curved adhesive breakage, such as... Figure 7 As shown: Determine the fifth pixel on each line segment formed by two edge pixels; If each fifth pixel point includes glue pixels, the glue path defect type is determined to exclude curved glue breakage. If each fifth pixel is a non-glue pixel, determine the third target pixel in each fifth pixel. The third target pixel is located on a straight line perpendicular to the glue path direction. Based on the preset adhesive path width, obtain two fourth target pixels that are located on the straight line and outside the edge of the adhesive path; Using a path search algorithm, a path search is performed starting from the third target pixel and ending at the two fourth target pixels. If the path search is successful, the type of adhesive defect in the adhesive path is determined to include curved adhesive breakage; If the path search fails, the type of adhesive defect in the determined adhesive path does not include curved adhesive breaks.

[0065] This application does not limit the position of the third target pixel. For example, the position of the third target pixel can be the center position on the line segment formed by the two edge pixels.

[0066] For example, the path search algorithm may employ the A-Star algorithm, the D-Star algorithm, or the best-first search algorithm.

[0067] The following is combined with Figure 8 The process of detecting curved adhesive breakage is explained. Figure 8 For explanations regarding each direction, please refer to [link / reference]. Figure 3 This will not be elaborated upon here. Figure 8 As shown, firstly, the fifth pixels on the line segment formed by edge pixels f and g are determined. Since each fifth pixel is a non-glue pixel, the third target pixel h among these fifth pixels is determined. Based on the preset glue path width, the fourth target pixel i1, located outside the pixel at glue path edge 1, and the fourth target pixel i2, located outside the pixel at glue path edge 2, are obtained. Using a path search algorithm, a path search is performed starting from the third target pixel h and ending at the fourth target pixels i1 and i2. Since the path search is successful, the glue path defect type is determined to include curved glue breakage.

[0068] In the example above, after determining the starting point (i.e. the third target pixel) and the ending point (i.e. the two fourth target pixels) of the path planning, the detection of irregularly shaped glue breaks, such as curved glue breaks, was achieved by using the path planning algorithm, thereby improving the detection capability of glue path defects.

[0069] In a feasible design, after detecting a curved glue breakage defect in the adhesive path, relevant information about the curved glue breakage is determined in the following way to facilitate subsequent research by the testing personnel: Two search paths are determined, one starting from the third target pixel and the other ending at the two fourth target pixels. Take the third target pixel as the first endpoint, and take the pixels adjacent to the first endpoint on each search path as the second endpoint, and determine the normals of the line segments formed by the first endpoint and each second endpoint. Starting from the first endpoint, search for non-glue pixels along both directions of each normal. If glue pixels are found along both directions of the normal, determine the two non-glue pixels located at the curved glue break edge that are adjacent to the glue pixels found. The second endpoint is used as the new first endpoint, and the pixel adjacent to the new first endpoint is used as the new second endpoint. The search is performed again to determine the two new non-glue pixels located at the edge of the curved glue break, until the glue pixel is not found in either direction of the normal within the third preset search length starting from the first endpoint. Based on the searched non-glue pixels located at the edge of the curved glue break, determine the two target non-glue pixels that are furthest apart; Determine the width and circumscribed rectangle of the curved glue break based on two target non-glue pixels.

[0070] The following is combined with Figure 9 The process for determining the relevant information regarding the curved breakage of the adhesive is explained above. For example... Figure 9 as well as Figure 8 As shown, Figure 8The search path starting from the third target pixel h and ending at the fourth target pixel i1 is path 1. The search path starting from the third target pixel h and ending at the fourth target pixel i2 is path 2. Taking path 1 as an example, the third target pixel h is taken as the first endpoint j1, and the pixel adjacent to the first endpoint j1 on path 1 is taken as the first endpoint j2. The normal k1 of the line segment formed by the first endpoint j1 and the first endpoint j2 is determined. Starting from the first endpoint j1, non-glue pixels are searched along the two directions of the normal. If glue pixels are found along the two directions of the normal, two non-glue pixels (m1 and m2) located on the curved glue break edge adjacent to the glue pixels are determined. The first endpoint j2 is taken as the new first endpoint j1, and the pixel adjacent to the new first endpoint j1 is taken as the new first endpoint j2. The search is performed again to determine the two new non-glue pixels located on the curved glue break edge. The search continues until no glue pixels are found along the normal in either direction within the third preset search length starting from the first endpoint, indicating that the glue edge has been reached. The search process for path 2 is described in the explanation of path 1 and will not be repeated here. Based on the non-glue pixels found on the curved glue break edge, the two farthest target non-glue pixels (n1 and n2) are determined. As shown in the diagram, the two farthest target non-glue pixels are the pixels at both ends of the curved glue break. The width of the curved glue break can be determined based on the difference in the x-coordinates of the two target non-glue pixels. This width is used as the length of the circumscribed rectangle, the preset glue path width is used as the width of the circumscribed rectangle, and the two target non-glue pixels are used as the two vertices of the circumscribed rectangle. The circumscribed rectangle can then be defined as follows: Figure 4 As shown.

[0071] The example above accurately identifies non-glue pixels located at the edge of a curved glue break by searching the normal of the search path. This accurately identifies the non-glue pixels at both ends of the curved glue break, allowing for the accurate and rapid determination of the width and circumscribed rectangle of the curved glue break based on these pixels.

[0072] Since the probability of beveled cut defects is lower than that of straight cut defects, and the probability of curved cut defects is lower than that of beveled cut defects, this application also provides... Figure 10 The step-by-step inspection method shown is used to detect defects in the adhesive path, thereby improving inspection efficiency.

[0073] like Figure 10 As shown, defects in the adhesive circuit are detected step by step through the following steps: Step a: Obtain a grayscale image containing the adhesive path to be detected; Step b: Determine the grayscale value of each first pixel within the target line in the grayscale image; Step c: Determine the edge pixels in the first pixel based on the grayscale value of each first pixel; Step d: Based on the edge pixel, determine whether the adhesive path defect type includes straight-edge glue breakage; if it includes, determine and output the defect information of straight-edge glue breakage; if it does not include, proceed to step e. Step e: Based on the edge pixel, determine whether the adhesive path defect type includes beveled adhesive breakage; if it includes, determine and output the defect information of beveled adhesive breakage; if it does not include, proceed to step f. Step f: Based on the edge pixel, determine whether the adhesive path defect type includes curved adhesive breakage; if it includes, determine and output the defect information of curved adhesive breakage; if it does not include, determine that there is no adhesive path defect.

[0074] For details on the specific implementation of each step, please refer to the descriptions of methods S110 to S140, which will not be repeated here.

[0075] In this embodiment, after acquiring a grayscale image containing the adhesive path to be detected, the grayscale value of each first pixel within the target path in the grayscale image is determined. To comprehensively detect all pixels in the adhesive path, the target path must penetrate the adhesive path along its direction. Based on the grayscale values ​​of each first pixel, if edge pixels can be identified, it indicates the presence of discontinuous adhesive areas within the adhesive path. Further analysis of these edge pixels can then determine whether adhesive path defects exist, thus achieving adhesive path detection.

[0076] Furthermore, edge pixels are pixels on the edges of areas without adhesive in the adhesive path. These areas may be caused by different types of adhesive path defects. The solution provided in this application can detect various types of adhesive path defects by detecting edge pixels.

[0077] In conjunction with the above embodiments of the adhesive circuit defect detection method, this application also provides an adhesive circuit defect detection device, such as... Figure 11 As shown, the device includes: The image acquisition module is used to acquire and send a grayscale image containing the adhesive path to be detected to the grayscale value determination module; The grayscale value determination module is used to determine and send the grayscale value of each first pixel in the target line in the grayscale image to the pixel determination module. The target line runs through the adhesive path along the adhesive path direction. The pixel determination module is used to determine and send the edge pixels in the first pixel to the glue path detection module based on the gray value of each first pixel. The difference between the gray value of the edge pixel and the gray value of the corresponding reference pixel is greater than a first threshold. The reference pixel corresponding to the edge pixel is located on the same target line as the edge pixel, and the edge pixel is adjacent to the corresponding reference pixel. The adhesive path detection module is used to determine whether there are adhesive path defects based on edge pixels.

[0078] In a feasible design, the pixel determination module determines whether there are adhesive defects in the adhesive path based on edge pixels in the following way: Determine each second pixel in each first pixel point, where the second pixel point is a non-glue pixel point located on the line segment formed by two edge pixels; Select two edge pixels and one pixel from each of the second pixels as the first target pixel; Based on the first target pixel, and according to the first preset search length, the search is performed along the first direction perpendicular to the adhesive path and the second direction perpendicular to the adhesive path, respectively. The searched pixels are the third pixels corresponding to the first target pixel. Based on the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel, determine whether the glue path defect type includes straight-edge glue breakage.

[0079] In a feasible design, the pixel determination module determines whether there are adhesive defects in the adhesive path based on edge pixels in the following way: Determine the center pixel of the line segment formed by two edge pixels; Determine each fourth pixel in each first pixel, where the fourth pixel is located on the line segment formed by the center pixel and the two edge pixels respectively; If each fourth pixel point includes glue pixels, the glue path defect type is determined to exclude the beveled glue breakage. If each fourth pixel is a non-glue pixel, determine two starting pixels located next to the center pixel along the first direction perpendicular to the glue path and the second direction perpendicular to the glue path, respectively. If both starting pixels are non-glue pixels, then starting from the two starting pixels, search for the second target pixel corresponding to the starting pixel along two directions parallel to the glue path. The second target pixel is a non-glue pixel, and the distance between the second target pixel and the starting point is within the second preset search length. After finding glue pixels within the second preset search length, a new center pixel is determined. The new center pixel is the center pixel of each second target pixel, and a new starting pixel corresponding to the new center pixel is determined. If all new starting pixels are non-glue pixels, starting from the new starting pixels, search for the second target pixels corresponding to the new starting pixels again along two directions parallel to the glue path, until no glue pixels are found within the second preset search length of the starting point. If all starting pixels and all second target pixels are non-adhesive pixels, the glue path defect type is determined to include beveled glue breakage. If each starting pixel and each second target pixel includes glue pixels, the glue path defect type is determined to exclude beveled glue breaks.

[0080] In a feasible design, the pixel determination module determines whether there are adhesive defects in the adhesive path based on edge pixels in the following way: Determine the fifth pixel on each line segment formed by two edge pixels; If each fifth pixel point includes glue pixels, the glue path defect type is determined to exclude curved glue breakage. If each fifth pixel is a non-glue pixel, determine the third target pixel in each fifth pixel. The third target pixel is located on a straight line perpendicular to the glue path direction. Based on the preset adhesive path width, obtain two fourth target pixels that are located on the straight line and outside the edge of the adhesive path; Using a path search algorithm, a path search is performed starting from the third target pixel and ending at the two fourth target pixels. If the path search is successful, the type of adhesive defect in the adhesive path is determined to include curved adhesive breakage; If the path search fails, the type of adhesive defect in the determined adhesive path does not include curved adhesive breaks.

[0081] In one feasible design, the device further includes a straight-edge glue breakage information acquisition module, which is used to determine the pixel width of the straight-edge glue breakage based on the pixel coordinates of two edge pixels if the glue path defect type includes straight-edge glue breakage.

[0082] In one feasible design, the straight-edge glue breakage information acquisition module is also used to determine the boundary for connected component analysis based on the pixel width, the preset pixel height of the glue path, and the pixel coordinates of the two edge pixels. The area of ​​the straight-cut glue break within the boundary is determined based on connected component analysis.

[0083] In one feasible design, the adhesive path detection module is also used to compare the second target pixel point located at the first end that was previously searched along the third direction parallel to the adhesive path with the second target pixel point located at the first end that was searched this time after each search in two directions parallel to the adhesive path direction, and to redetermine the second target pixel point located at the first end based on the comparison result. After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the second end that was previously searched in the fourth direction parallel to the adhesive path is compared with the second target pixel point located at the second end that was searched in this search. The second target pixel point located at the second end is re-determined based on the comparison result. The first direction and the second direction are opposite directions.

[0084] In one feasible design, the device also includes a beveled glue breakage information acquisition module, which is used to acquire the sum of the number of each second target pixel point searched in each search process, and determine the sum as the area of ​​the beveled glue breakage. Alternatively, the number of center pixels found can be determined as the height of the beveled glue cut. Alternatively, the circumscribed rectangle of the angled glue cut can be determined based on the second target pixel at the first end and the second target pixel at the second end.

[0085] In one feasible design, the device also includes a curved adhesive breakage information acquisition module, used for... Two search paths are determined, one starting from the third target pixel and the other ending at the two fourth target pixels. Take the third target pixel as the first endpoint, and take the pixels adjacent to the first endpoint on each search path as the second endpoint, and determine the normals of the line segments formed by the first endpoint and each second endpoint. Starting from the first endpoint, search for non-glue pixels along both directions of each normal. If glue pixels are found along both directions of the normal, determine the two non-glue pixels located at the curved glue break edge that are adjacent to the glue pixels found. The second endpoint is used as the new first endpoint, and the pixel adjacent to the new first endpoint is used as the new second endpoint. The search is performed again to determine the two new non-glue pixels located at the edge of the curved glue break, until the glue pixel is not found in either direction of the normal within the third preset search length starting from the first endpoint. Based on the searched non-glue pixels located at the edge of the curved glue break, determine the two target non-glue pixels that are furthest apart; Determine the width and circumscribed rectangle of the curved glue break based on two target non-glue pixels.

[0086] Other implementation methods and effects of this device are described in the embodiments of the adhesive circuit defect detection method, and will not be repeated here.

[0087] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0088] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0089] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0090] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0091] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0092] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for detecting defects in adhesive circuits, characterized in that, include: Obtain a grayscale image containing the adhesive path to be detected; Determine the grayscale value of each first pixel point within the target line in the grayscale image, wherein the target line penetrates the adhesive path along the adhesive path direction; Based on the grayscale value of each first pixel, edge pixels in the first pixel are determined. The difference between the grayscale value of each edge pixel and the grayscale value of the corresponding reference pixel is greater than a first threshold. The reference pixel corresponding to the edge pixel is located on the same target line as the edge pixel, and the edge pixel is adjacent to the corresponding reference pixel. Based on the edge pixels, determine whether there are adhesive defects in the adhesive path; The step of determining whether there are adhesive defects in the adhesive path based on the edge pixels includes: Determine each second pixel in each of the first pixels, wherein the second pixel is a non-glue pixel located on the line segment formed by two edge pixels; Select two edge pixels and one pixel from each of the second pixels as the first target pixel; Based on the first target pixel, and according to the first preset search length, a search is performed along a first direction perpendicular to the adhesive path and a second direction perpendicular to the adhesive path, respectively. Each pixel found is a third pixel corresponding to the first target pixel. Based on the difference between the gray value of each first target pixel and the gray value of each corresponding third pixel, it is determined whether the glue path defect type includes straight-edge glue breakage.

2. The method according to claim 1, characterized in that, The step of determining whether there are adhesive defects in the adhesive path based on the edge pixels includes: Determine the center pixel of the line segment formed by the two edge pixels; Determine each fourth pixel in each of the first pixels, wherein the fourth pixel is located on the line segment formed by the center pixel and the two edge pixels respectively; If each of the fourth pixels includes glue pixels, the glue path defect type of the glue path is determined to not include the angled glue break. If each of the fourth pixels is a non-adhesive pixel, two starting pixels located at the next position of the center pixel are determined along a first direction perpendicular to the adhesive path and a second direction perpendicular to the adhesive path. If both starting pixels are non-glue pixels, then starting from the two starting pixels, search for the second target pixel corresponding to the starting pixel along two directions parallel to the glue path. The second target pixel is a non-glue pixel, and the distance between the second target pixel and the starting point is within the second preset search length. After finding glue pixels within the second preset search length, a new center pixel is determined. The new center pixel is the center pixel of each of the second target pixels, and a new starting pixel is determined corresponding to the new center pixel. If all the new starting pixels are non-glue pixels, starting from the new starting pixels, search for the second target pixels corresponding to the new starting pixels again along two directions parallel to the glue path, until no glue pixels are found within the second preset search length at the starting point. If each of the starting pixels and each of the second target pixels are non-adhesive pixels, the adhesive path defect type of the adhesive path is determined to include oblique glue breakage; If each of the starting pixels and each of the second target pixels includes glue pixels, the glue path defect type of the glue path is determined to not include beveled glue breakage.

3. The method according to claim 1, characterized in that, The step of determining whether there are adhesive defects in the adhesive path based on the edge pixels includes: Determine the fifth pixel point on each line segment formed by the two edge pixels; If each of the fifth pixels includes a glue pixel, then the glue path defect type of the glue path is determined to not include curved glue breakage; If all the fifth pixels are non-adhesive pixels, determine the third target pixel among the fifth pixels, and the third target pixel is located on a straight line perpendicular to the direction of the adhesive path; Based on the preset adhesive path width, obtain two fourth target pixels located on the straight line and outside the edge of the adhesive path; Using a path search algorithm, a path search is performed starting from the third target pixel and ending at the two fourth target pixels; If the path search is successful, the type of adhesive defect in the adhesive path is determined to include curved adhesive breakage; If the path search fails, the type of adhesive defect in the adhesive path is determined to be excluding curved adhesive breaks.

4. The method according to claim 1, characterized in that, If the adhesive path defect type includes straight-edge breakage, the method further includes: The pixel width of the straight-cut adhesive break is determined based on the pixel coordinates of the two edge pixels.

5. The method according to claim 4, characterized in that, If the adhesive path defect type includes straight-edge breakage, the method further includes: The boundary for connected component analysis is determined based on the pixel width, the preset pixel height of the adhesive path, and the pixel coordinates of the two edge pixels. The area of ​​the straight-cut glue break within the boundary is determined based on connected component analysis.

6. The method according to claim 2, characterized in that, Also includes: After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the first end that was previously searched in a third direction parallel to the adhesive path is compared with the second target pixel point located at the first end that was searched in this search, and the second target pixel point located at the first end is re-determined based on the comparison result. After each search is completed in two directions parallel to the adhesive path, the second target pixel point located at the second end that was previously searched in the fourth direction parallel to the adhesive path is compared with the second target pixel point located at the second end that was searched in this search, and the second target pixel point located at the second end is re-determined based on the comparison result. The first direction and the second direction are opposite directions.

7. The method according to claim 6, characterized in that, After determining that the type of adhesive defect in the adhesive path includes beveled adhesive breakage, the method further includes: The sum of the number of each second target pixel found in each search process is obtained, and the sum is determined as the area of ​​the angled glue break. Alternatively, the number of center pixels found can be determined as the height of the beveled glue cut. Alternatively, the outer rectangle of the beveled glue cut can be determined based on the second target pixel located at the first end and the second target pixel located at the second end.

8. The method according to claim 3, characterized in that, If the adhesive path defect type includes curved adhesive breakage, the method further includes: Two search paths are determined, one starting from the third target pixel and the other ending at the two fourth target pixels. Using the third target pixel as the first endpoint and the pixels adjacent to the first endpoint on each search path as the second endpoint, determine the normals of the line segments formed by the first endpoint and each of the second endpoints. Starting from the first endpoint, search for non-glue pixels along the two directions of each normal. If glue pixels are found along the two directions of the normal, determine two non-glue pixels located at the curved glue break edge that are adjacent to the glue pixels found. The second endpoint is used as the new first endpoint, and the pixel adjacent to the new first endpoint is used as the new second endpoint. The search is performed again to determine two new non-glue pixels located at the edge of the curved glue break, until no glue pixels are found in either direction along the normal within a third preset search length starting from the first endpoint. Based on the searched non-glue pixels located at the edge of the curved glue break, determine the two target non-glue pixels that are furthest apart; The width and circumscribed rectangle of the curved glue break are determined based on two target non-glue pixels.

9. A device for detecting defects in adhesive circuits, the device being applicable to the method described in any one of claims 1 to 8, characterized in that, include: The image acquisition module is used to acquire and send a grayscale image containing the adhesive path to be detected to the grayscale value determination module; The grayscale value determination module is used to determine and send the grayscale value of each first pixel in the target line in the grayscale image to the pixel point determination module. The target line runs through the adhesive path along the adhesive path direction. The pixel determination module is used to determine and send edge pixels in the first pixels to the adhesive path detection module based on the gray value of each first pixel. The difference between the gray value of each edge pixel and the gray value of the corresponding reference pixel is greater than a first threshold. The reference pixel corresponding to the edge pixel is located on the same target line as the edge pixel, and the edge pixel is adjacent to the corresponding reference pixel. The adhesive path detection module is used to determine whether there are adhesive path defects based on the edge pixels.

Images