Grid line identification method of solar cell panel

Abstract

The invention relates to the technical field of data identification, in particular to a solar cell panel grid line identification method, which comprises the following steps: acquiring a surface gray level image of a solar cell panel; obtaining a template grid line and window areas on the two sides of the template grid line; obtaining a clustering cluster and a pixel vacancy rate of the clustering cluster; identifying a final grid line to be detected; and judging whether the final to-be-detected grid line is abnormal or not. According to the invention, the clear final to-be-detected grid line is identified, and the influence of illumination on grid line extraction is avoided, so that the accuracy of grid line anomaly detection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data recognition, and in particular to a grid line recognition method for a solar cell panel. Background Art

[0002] As the concepts of low-carbon, environmental protection and green development become more and more deeply rooted, photovoltaic power generation technology has been widely used. Photovoltaic power generation technology mainly uses solar panels as the key components of photovoltaic power generation equipment. Because solar panels are exposed to outdoor environments for a long time, they have strict quality requirements, especially the grid lines of solar panels. Since the grid lines of solar panels are mainly used as circuits, it is very important to detect whether the distribution of the grid lines of solar panels is uniform and whether there are any abnormalities.

[0003] In the prior art, when detecting the grid lines of a solar panel, the Hough line detection is mainly performed to obtain the grid line area. However, in the process of using the prior art, the surface of the solar panel is prone to reflection, that is, the grid lines are disturbed by light, which will cause the grid lines to be unclear and unable to be accurately identified. Summary of the invention

[0004] The present invention provides a grid line recognition method for a solar cell panel, so as to solve the problem that the existing method cannot accurately match and recognize the grid lines, and thus cannot accurately judge whether the grid lines are abnormal.

[0005] A grid line identification method of a solar panel of the present invention adopts the following technical solution:

[0006] Get the surface grayscale image of the solar panel;

[0007] Obtain bright spots in the surface grayscale image and use the bright spots as grid line positioning points, and obtain the template grid lines of the solar panel according to the grid line positioning points;

[0008] A sliding window symmetrical to the template grid line is set on both sides of the template grid line, wherein the sliding window starts from the grid line positioning point and slides along the direction of the template grid line, and a window area after each sliding is obtained;

[0009] Obtaining the pixel vacancy rate of each window area according to the number of vacant pixel points in the window area; performing hierarchical clustering on the pixel vacancy rates of the window areas on both sides of each template grid line to obtain multiple clusters, and obtaining the pixel vacancy rate of the cluster according to the pixel vacancy rates of all window areas in each cluster;

[0010] According to the pixel vacancy rate of each cluster and the difference between the pixel vacancy rates of every two window areas symmetrical about the template grid line in the cluster, the irregularity degree of the edge of the to-be-detected grid line corresponding to each cluster is obtained;

[0011] According to the irregularity of the edge of the to-be-detected grid line corresponding to the cluster, the size of the structural element of the cluster is obtained, and the area corresponding to the size of the structural element of the cluster is morphologically eroded to obtain the final to-be-detected grid line.

[0012] Preferably, obtaining the pixel vacancy rate of each window area includes:

[0013] Get the grayscale mean of all vacant pixels in the window area;

[0014] The grayscale mean of all vacant pixels in the window area is taken as the pixel vacancy rate of the window area.

[0015] Preferably, obtaining the pixel vacancy rate of the cluster includes:

[0016] The average of the pixel vacancy rates of all window areas within the cluster is taken as the pixel vacancy rate of the cluster.

[0017] Preferably, obtaining the irregularity degree of the edge of the to-be-detected grid line corresponding to each cluster includes:

[0018] Obtain the absolute value of the difference in pixel vacancy rate between every two window areas symmetrical about the template grid line within the cluster;

[0019] The sum of the absolute values ​​of the differences of the pixel vacancy rates of two window areas symmetrical about the template grid line in the cluster is multiplied by the pixel vacancy rate of the cluster to obtain the first target value;

[0020] The first target value is normalized and calculated to obtain the irregularity of the edge of the to-be-detected grid line corresponding to the cluster.

[0021] Preferably, obtaining the size of the structural element of the cluster includes:

[0022] The irregularity of the edge of the to-be-detected grid line corresponding to the cluster is multiplied by a preset threshold value to obtain a second target value;

[0023] The second target value is rounded up to obtain the side length of the structural element of the corresponding cluster;

[0024] The size of the structural element of the cluster is obtained according to the side length of the structural element of the cluster.

[0025] Preferably, obtaining bright spots in the surface grayscale image includes:

[0026] Perform threshold segmentation on the surface grayscale image to obtain grid line pixel points;

[0027] Performing edge detection on the grid line pixel points to obtain the grid line to be detected;

[0028] The intersection of the grid lines to be detected is the bright spot in the surface grayscale image.

[0029] Preferably, obtaining a template grid line of a solar panel includes:

[0030] Two mutually perpendicular straight lines are set on the surface grayscale image, wherein one of the straight lines is perpendicular to the edge of the surface grayscale image;

[0031] Move along the horizontal direction or vertical direction of the surface grayscale image respectively, and when the straight line passes through the grid line positioning point, the straight line is used as a template grid line;

[0032] Traverse and slide in sequence until all grid line positioning points are on a straight line, that is, all template grid lines of the solar panel are obtained.

[0033] Preferably, the template grid line is a single pixel line.

[0034] Preferably, pixels within the window area whose grayscale values ​​are greater than the grayscale mean value corresponding to the window area are regarded as vacant pixels.

[0035] The beneficial effects of the grid line identification method of a solar panel of the present invention are:

[0036] Through the analysis of the structure of the grid lines and single crystals of the solar cell panel, the grid lines form intersections, and the grid lines and the grid lines surround the single crystal area, so the bright spots in the surface grayscale image are first obtained and the bright spots are used as grid line positioning points to obtain the template grid lines of the solar cell panel. Due to the influence of light, it is difficult to directly obtain the edge of the grid line to be detected. Therefore, in order to obtain the edge of the grid line to be detected, the area where the edge of the grid line to be detected is located is first narrowed to improve the detection efficiency, that is, sliding windows symmetrical about the template grid line are set on both sides of the template grid line, wherein the sliding window starts from the grid line positioning point and slides along the direction of the template grid line, and the window area after each sliding is obtained to ensure that the edge of the grid line to be detected Located in the window area, then, based on the characteristics that the grayscale of the grid line and the surface of the single wafer are different, the vacant pixel points are obtained, so as to obtain the pixel vacancy rate of each window area according to the number of vacant pixel points in the window area, and then the window area is hierarchically clustered using the characteristics of the window areas with similar pixel vacancy rates, so as to obtain the pixel vacancy rate of the cluster cluster, and then based on the pixel vacancy rate of each cluster cluster, the grid line to be detected in the window area is accurately extracted, that is, the irregularity of the edge of the grid line to be detected is obtained according to the pixel vacancy rate of the cluster clusters piled up on both sides of the template grid line, so as to determine the size of the structural element according to the irregularity of the edge of the grid line to be detected, and then the final grid line to be detected is obtained. The present invention realizes the recognition of the final grid line to be detected with clarity. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 The present invention is a flow chart of an embodiment of a method for identifying grid lines of a solar panel. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] An embodiment of a solar panel grid line identification method of the present invention is used in the following scenarios: in the production and assembly of solar panels, the grid lines of solar panels are detected to determine whether the single wafers of the solar panels are installed intact, so as to prevent the single wafers from being skewed or warped, which may lead to poor connection between the single wafers and the panel body, thereby affecting the power generation function of the solar panel. Figure 1 As shown, this embodiment includes:

[0041] S1, obtaining a surface grayscale image of a solar panel;

[0042] Specifically, this embodiment mainly detects the grid lines of the TOPCon structure solar panel (tunneling oxide passivation contact structure solar panel), wherein this embodiment uses a high-definition camera to collect the surface image of the solar panel, and then grayscales the collected surface image to obtain a surface grayscale image of the solar panel, wherein the grayscale processing of the image is a prior art and will not be repeated in this embodiment.

[0043] S2, obtaining the template grid line and the window areas on both sides of the template grid line;

[0044] Specifically, bright spots in the surface grayscale image are obtained and used as grid line positioning points, and the template grid lines of the solar panel are obtained according to the grid line positioning points; sliding windows symmetrical about the template grid lines are set on both sides of the template grid lines, wherein the sliding window slides along the direction of the template grid lines starting from the grid line positioning points, and the window area after each sliding is obtained. It should be noted that one of the sides along the sliding direction of the sliding window is colinear with the template grid line.

[0045] Since the single crystals laid on the surface of the solar panel are arranged neatly, there will be a certain distance between each single crystal. Because there are no crystals laid on the corners of the single crystal, the metal on the bottom of the board will be exposed to form bright spots. Therefore, the position of the bright spots represents the distribution of the single crystals. By obtaining the bright spots in the image as the positioning points of the grid lines, and then fitting the template grid lines according to the bright spots, the subsequent grid lines obtained according to the image segmentation are the actual grid lines. Therefore, the bright spots in the surface grayscale image are first obtained and used as the grid line positioning points, and the template grid lines of the solar panel are obtained according to the grid line positioning points.

[0046] Among them, obtaining the bright spots in the surface grayscale image includes threshold segmentation of the surface grayscale image to obtain grid line pixel points; edge detection of the grid line pixel points to obtain the grid line to be detected; the intersection of the grid line to be detected is the bright spot in the surface grayscale image. Specifically: because the grayscale of the grid line and the surface of the single crystal chip is different, the grayscale value of the grid line is larger, and the grayscale value of the single crystal chip is smaller. Therefore, the threshold of the threshold segmentation algorithm in the embodiment itself is set to 150, and the grid line is separated from the single crystal chip by threshold segmentation, because the grayscale difference between the white straight line and other areas in the grayscale image is obvious, and what is to be segmented is the white line, that is, the grid line. This threshold is an empirical threshold, and the implementer can adjust it according to the specific situation. Self-setting, that is, pixel points with grayscale values ​​greater than the preset threshold are used as pixel points of the gate line, and pixel points less than or equal to the preset threshold are used as pixel points of the single chip. At this point, the gate line pixel points are obtained. Because the bright spot in the grayscale image is at the junction of the gate lines, the position of the bright spot is obtained according to the distribution of the gate lines. Here, the obtained gate line pixel points are subjected to Canny operator detection (edge ​​detection) to obtain the straight lines where all the gate lines in the image are located. The Canny operator detection algorithm is an existing well-known technology and will not be repeated here. Then, the intersection points of all the straight lines are obtained and marked. The intersection points are the bright spots in the image, and the bright spots are used as the gate line positioning points.

[0047] Among them, obtaining the template grid lines of the solar panel includes: setting two mutually perpendicular straight lines on the surface grayscale image, one of which is perpendicular to the edge of the surface grayscale image; moving along the horizontal direction or vertical direction of the surface grayscale image respectively, and when the straight line passes through the grid line positioning point, the straight line is used as a template grid line; traversing and sliding in sequence until all the grid line positioning points are on the straight line, that is, all the template grid lines of the solar panel are obtained, wherein it should be noted that the template grid line is a single-pixel line.

[0048] Secondly, due to the influence of light, some grid lines on the solar panel are not clear, which may form brighter spots, resulting in the directly obtained grid lines not being the real grid lines. However, under similar light gain, the original grid lines are still brighter than the surrounding ones, but the difference becomes smaller, which makes direct detection prone to large errors. Therefore, it is necessary to gradually obtain the real grid lines through morphological corrosion, and it is impossible to directly obtain the grid line area. Therefore, it is necessary to first set a sliding window symmetrical about the template grid line on both sides of the template grid line, where the sliding window starts from the grid line positioning point and slides along the direction of the template grid line, and obtains the window area after each sliding. For the vertical template grid line, the corresponding window area distribution On the left and right sides of the vertical template grid line, and for the horizontal template grid line, the corresponding window area is distributed on the upper and lower sides of the horizontal template grid line, that is, the specific acquisition window area includes: when searching through the preset template grid line, taking the vertical template grid line as an example, the left and right sides of the vertical template grid line are single crystals, showing a symmetrical distribution. Therefore, when performing pixel search, the left and right pixels are complemented according to the changes of the pixel points on the left and right sides of the vertical template grid line, and then the pixel points at the vacant position are obtained. Therefore, a 5*5 sliding window is set, starting from the positioning point, and sliding along the left and right sides of the vertical template grid line respectively to obtain a 5*5 window area.

[0049] It should be noted that the purpose of obtaining the window area is that the gate line to be detected must be along the template gate line, so the window area is obtained on both sides of the template gate line, so that the edge of the gate line to be detected is located in the window area, so as to facilitate the subsequent accurate acquisition of the edge line to be detected. In order to make the edge of the gate line to be detected located in the window area, it is necessary to set a larger sliding window size, so this embodiment takes an empirical value of 5*5 sliding window.

[0050] S3, obtaining clusters and taking pixel vacancy rates of the clusters;

[0051] After obtaining the gate line positioning point, it is necessary to search for pixels according to the position of the gate line positioning point to obtain the window area, that is, it is obtained that the boundary of the gate line to be detected is located in the window area, and the window area contains not only pixel points corresponding to the single chip but also pixel points of the gate line to be detected. For the window area, most of the pixels therein are corresponding to the single chip. Therefore, in this embodiment, it is necessary to first identify the non-single chip pixels in the window area, and then cluster them based on the proportion of the non-single chip pixels, and obtain the pixel vacancy rate of the cluster cluster. Therefore, based on the window area obtained in step S2, the pixel vacancy rate of each window area is obtained according to the number of vacant pixels in the window area in the window area; the pixel vacancy rates of the window areas on both sides of each template gate line are hierarchically clustered to obtain multiple cluster clusters, and the pixel vacancy rate of the cluster cluster is obtained according to the pixel vacancy rates of all window areas in each cluster cluster. Among them, because hierarchical clustering is clustering a continuous change sequence, there will be no jump. Hierarchical clustering is a prior art and this embodiment will not be repeated.

[0052] Wherein, the pixel point whose gray value in the window area is greater than the gray mean value corresponding to the window area is regarded as a vacant pixel point, that is, the gray value of the pixel point in the window area is compared with the average value, if the gray value of the pixel point is greater than the average gray value, it means that the pixel point is not a pixel point in the single-chip area, then the pixel point is regarded as a vacant pixel point, and is assigned a value of 1, and when the gray value of the pixel point is less than the average gray value, it means that the pixel point is a pixel point in the single-chip area, therefore, the pixel point is not a vacant pixel point, and is assigned a value of 0; wherein, obtaining the pixel vacancy rate of each window area includes: obtaining the gray mean value of all vacant pixel points in the window area; taking the gray mean value of all vacant pixel points in the window area as the pixel vacancy rate of the window area, that is, the calculation formula of the pixel vacancy rate of the window area is:

[0053]

[0054] Where D j represents the pixel vacancy rate of the jth window area;

[0055] I i Represents the gray value of the i-th vacant pixel in the j-th window area;

[0056] m represents the total number of vacant pixels in the jth window area;

[0057] It should be noted that during the sliding of the window, the vacancy rate of the pixel points is obtained by comparing the grayscale changes of the pixel points in the window area on both sides of the template gate line. Because the grayscale value of the pixel points at the gate line is large, and the grayscale value of the pixel points at the single chip is small, if the gate line is bent, then in the process of sliding along the template gate line, the grayscale of the pixel points on the left and right sides will be different, so the pixel vacancy rate is different. Therefore, the pixel vacancy rate is used to represent the change of the pixel points of the actual gate line in each window area.

[0058] The pixel vacancy rate of the cluster is obtained, including: taking the average of the pixel vacancy rates of all window areas in the cluster as the pixel vacancy rate of the cluster. The calculation formula of the pixel vacancy rate of the cluster is:

[0059]

[0060] Where D r ′ represents the pixel vacancy rate of the rth cluster;

[0061] n represents the number of window areas in the rth cluster;

[0062] D j represents the pixel vacancy rate of the jth window area;

[0063] It should be noted that hierarchical clustering will group window areas with similar pixel vacancy rates into one category. Therefore, the larger the average pixel vacancy rate, the larger the pixel vacancy rate of the cluster.

[0064] S4, obtaining the final gate line to be detected;

[0065] Based on step S2, edge lines are searched in the direction of the four template grid lines connected to the grid line positioning point to obtain the fuzzy edge of the single chip, that is, the edge to be detected of the single chip is located in the window area, and then the fuzzy edge is accurately obtained in combination with the pixel vacancy rate of each cluster cluster obtained in step S3. Therefore, the irregularity of the edge of the grid line to be detected corresponding to each cluster cluster is obtained according to the pixel vacancy rate of each cluster cluster and the difference in the pixel vacancy rate of each two window areas symmetrical about the template grid line in the cluster cluster; according to the irregularity of the edge of the grid line to be detected corresponding to the cluster cluster, the size of the structural element of the cluster cluster is obtained, and the area corresponding to the size of the structural element of the cluster cluster is morphologically eroded to obtain the final grid line to be detected.

[0066] Specifically, obtaining the degree of irregularity of the edge of the to-be-detected grid line corresponding to each clustering cluster includes: obtaining the absolute value of the difference of the pixel vacancy rate of each two window areas symmetrical about the template grid line in the clustering cluster; multiplying the sum of the absolute values ​​of the difference of the pixel vacancy rate of all two window areas symmetrical about the template grid line in the clustering cluster by the pixel vacancy rate for the clustering cluster to obtain a first target value; performing normalization calculation on the first target value to obtain the degree of irregularity of the edge of the to-be-detected grid line corresponding to the clustering cluster, wherein the calculation formula of the degree of irregularity of the edge of the to-be-detected grid line corresponding to the clustering cluster is:

[0067]

[0068] In the formula, K r Indicates the irregularity of the edge of the grid line to be detected corresponding to the rth cluster;

[0069] D r ′ represents the pixel vacancy rate of the rth cluster;

[0070] D jz represents the pixel vacancy rate of the window area z in the jth group of symmetrical window areas about the template grid line;

[0071] D jy represents the pixel vacancy rate of the window area y in the jth group of symmetrical window areas about the template grid line;

[0072] N represents the total number of window regions symmetrical about the template grid line in the rth cluster;

[0073] Norm represents a normalization function, the purpose of which is to make the value range of the irregularity of the edge of the gate line to be detected be [0,1];

[0074] It should be noted that |D jz -D jy | represents the difference of pixel points in the window area on both sides of the template grid line. Therefore, when performing according to the preset grid line, if there is no abnormality, the difference between the two sides of the grid line is 0. When the grid line is abnormal, the difference is not 0, and the greater the abnormality of the grid line, the greater the difference. Therefore, the difference of pixel vacancy rates in all window areas in the cluster is calculated. The greater the difference in pixel vacancy rates, the greater the irregularity of the edge of the grid line to be detected. Multiply it by the pixel vacancy rate D of the cluster. r ′ is because in the same cluster, the pixel vacancy rate in each window area is similar. Therefore, the pixel vacancy rate of the obtained cluster can represent the overall pixel change of the cluster. Therefore, the irregularity of the edge of the gate line to be detected located in different window areas is calculated according to the vacancy rate of each cluster. Based on this, the irregularity of different window areas is obtained, that is, the irregularity of the local edge of the actual edge line in the corresponding cluster.

[0075] Since the greater the degree of irregularity of the edge of the to-be-detected grid line, the greater the degree of abnormality of the to-be-detected grid line, when the corrosion operation is performed on the area corresponding to the to-be-detected grid line, a larger-sized structural element is required to obtain a better processing effect, so that the empty pixel points can be filled to a certain extent, that is, the morphological corrosion is mainly to determine the size of the structural element. If the size is too small, the complete edge cannot be obtained. If the size is too large, the obtained edge range will be expanded. Therefore, it is necessary to obtain the size of the structural element of the cluster cluster according to the degree of irregularity of the edge of the to-be-detected grid line corresponding to each cluster cluster obtained. Specifically, obtaining the size of the structural element of the cluster cluster includes: multiplying the degree of irregularity of the edge of the to-be-detected grid line corresponding to the cluster cluster by a preset threshold value to obtain a second target value; rounding up the second target value to obtain the side length of the structural element of the corresponding cluster cluster; obtaining the size of the structural element of the cluster cluster according to the side length of the structural element of the cluster cluster, and the calculation formula for the side length of the structural element of the cluster cluster is:

[0076]

[0077] In the formula, w r represents the side length of the structural element corresponding to the rth cluster. The size of the structural element can be obtained based on the side length of the structural element.

[0078] Indicates the rounding up symbol;

[0079] K r Indicates the irregularity of the edge of the grid line to be detected corresponding to the rth cluster;

[0080] It should be noted that 10 represents the preset threshold, which is multiplied by 10 because the irregularity level K r The range is [0,1], and the size of the structure element is a positive integer, so it is enlarged tenfold and then rounded.

[0081] Finally, the area corresponding to the size of the structural element of the cluster is morphologically eroded to obtain the final grid line to be detected, and the final grid line to be detected is a clear grid line to be detected.

[0082] Specifically, the degree of irregularity of the final gate line to be detected is obtained according to the distance between the edge of the final gate line to be detected and the corresponding template gate line, and whether the final gate line to be detected has an abnormality is determined according to the degree of irregularity of the final gate line to be detected.

[0083] The final irregularity of the gate line to be detected includes: obtaining the total number difference between the pixel points of the actual single-chip region where the final gate line to be detected is located and the total number difference between the pixel points of the template single-chip region where the template gate line corresponding to the final gate line to be detected is located; normalizing the total number difference between each actual single-chip region and the corresponding template single-chip region to obtain the irregularity of the final gate line to be detected corresponding to the actual single-chip region, that is, the calculation formula of the irregularity of the final gate line to be detected is:

[0084] Q=Norm[T rm -T rd ]

[0085] In the formula, Q represents the degree of irregularity of the final gate line to be detected corresponding to the actual single wafer area;

[0086] T rm Indicates the total number of pixels in the actual single wafer area where the final gate line to be detected is located;

[0087] T rd Indicates the total number of pixel points in the template single wafer region where the template grid line corresponding to the final grid line to be detected is located;

[0088] Norm represents a normalization function, the purpose of which is to make the value range of the irregularity of the final gate line to be detected corresponding to the actual single-wafer area be [0,1], so as to facilitate the setting of subsequent thresholds.

[0089] It should be noted that [T rm -T rd ] represents the total number difference between the pixel points of the actual single-chip area where the final gate line to be detected is located and the total number of pixel points of the template single-chip area where the template gate line corresponding to the final gate line to be detected is located. The larger the total number difference is, the greater the degree of irregularity of the final gate line to be detected corresponding to the actual single-chip area is.

[0090] Among them, judging whether there is an abnormality in the final gate line to be detected includes: based on the obtained irregularity degree of the final gate line to be detected corresponding to the actual single-chip area, and the value range of the irregularity degree of the final gate line to be detected is [0,1], so the regularity threshold is set to 0.12. When the irregularity degree of the final gate line to be detected corresponding to the actual single-chip area is greater than 0.12, it indicates that there is an abnormality in the final gate line to be detected corresponding to the actual single-chip area; when the irregularity degree of the final gate line to be detected corresponding to the actual single-chip area is less than or equal to 0.12, it indicates that there is no abnormality in the final gate line to be detected corresponding to the actual single-chip area.

[0091] The invention discloses a method for identifying grid lines of solar panels. The method analyzes the structure of grid lines and single wafers of solar panels. Intersections are formed between grid lines. Grid lines and single wafer areas are surrounded by grid lines. Therefore, bright spots in a surface grayscale image are first obtained and used as grid line positioning points to obtain template grid lines of solar panels. Due to the influence of light, it is difficult to directly obtain the edge of the grid line to be detected. Therefore, in order to obtain the edge of the grid line to be detected, the area where the edge of the grid line to be detected is located is first narrowed to improve the detection efficiency. That is, sliding windows symmetrical about the template grid line are set on both sides of the template grid line. The sliding window starts from the grid line positioning point and slides along the direction of the template grid line. The window area after each sliding is obtained to ensure that the grid line to be detected is The edge of the line is located in the window area, and then, based on the different grayscale characteristics of the gate line and the single chip surface, the vacant pixel points are obtained, and the pixel vacancy rate of each window area is obtained according to the number of vacant pixel points in the window area, and then the window area is hierarchically clustered using the characteristics of the window areas with similar pixel vacancy rates, so as to obtain the pixel vacancy rate of the cluster cluster, and then based on the pixel vacancy rate of each cluster cluster, the gate line to be detected in the window area is accurately extracted, that is, the irregularity of the edge of the gate line to be detected is obtained according to the pixel vacancy rate of the cluster clusters stacked on both sides of the template gate line, so as to determine the size of the structural element according to the irregularity of the edge of the gate line to be detected, and then the final gate line to be detected is obtained, thereby achieving accurate gate line recognition.

Claims

1. A method for identifying grid lines of a solar panel, characterized in that: include: Get the surface grayscale image of the solar panel; Obtain bright spots in the surface grayscale image and use the bright spots as grid line positioning points, and obtain the template grid lines of the solar panel according to the grid line positioning points; A sliding window symmetrical to the template grid line is set on both sides of the template grid line, wherein the sliding window starts from the grid line positioning point and slides along the direction of the template grid line, and a window area after each sliding is obtained; Obtaining the pixel vacancy rate of each window area according to the number of vacant pixel points in the window area; performing hierarchical clustering on the pixel vacancy rates of the window areas on both sides of each template grid line to obtain multiple clusters, and obtaining the pixel vacancy rate of the cluster according to the pixel vacancy rates of all window areas in each cluster; According to the pixel vacancy rate of each cluster and the difference between the pixel vacancy rates of every two window areas symmetrical about the template grid line in the cluster, the irregularity degree of the edge of the to-be-detected grid line corresponding to each cluster is obtained; According to the irregularity of the edge of the to-be-detected grid line corresponding to the cluster, the size of the structural element of the cluster is obtained, and the area corresponding to the size of the structural element of the cluster is morphologically eroded to obtain the final to-be-detected grid line.

2. A method for identifying grid lines of a solar panel according to claim 1, characterized in that: Get the pixel vacancy rate of each window area, including: Get the grayscale mean of all vacant pixels in the window area; The grayscale mean of all vacant pixels in the window area is taken as the pixel vacancy rate of the window area.

3. A grid line identification method for a solar panel according to claim 1, characterized in that: Get the pixel vacancy rate of the cluster, including: The average of the pixel vacancy rates of all window areas within the cluster is taken as the pixel vacancy rate of the cluster.

4. A method for identifying grid lines of a solar panel according to claim 1, characterized in that: Obtain the irregularity of the edge of the to-be-detected grid line corresponding to each cluster, including: Obtain the absolute value of the difference in pixel vacancy rate between every two window areas symmetrical about the template grid line within the cluster; The sum of the absolute values ​​of the differences of the pixel vacancy rates of two window areas symmetrical about the template grid line in the cluster is multiplied by the pixel vacancy rate of the cluster to obtain the first target value; The first target value is normalized and calculated to obtain the irregularity of the edge of the to-be-detected grid line corresponding to the cluster.

5. The method for identifying grid lines of a solar panel according to claim 1, characterized in that: Get the size of the structural element of the cluster, including: The irregularity of the edge of the to-be-detected grid line corresponding to the cluster is multiplied by a preset threshold value to obtain a second target value; The second target value is rounded up to obtain the side length of the structural element of the corresponding cluster; The size of the structural element of the cluster is obtained according to the side length of the structural element of the cluster.

6. A method for identifying grid lines of a solar panel according to claim 1, characterized in that: Get the highlights in the surface grayscale image, including: Perform threshold segmentation on the surface grayscale image to obtain grid line pixel points; Performing edge detection on the grid line pixel points to obtain the grid line to be detected; The intersection of the grid lines to be detected is the bright spot in the surface grayscale image.

7. A method for identifying grid lines of a solar panel according to claim 1, characterized in that: Get the template grid lines of the solar panel, including: Two mutually perpendicular straight lines are set on the surface grayscale image, wherein one of the straight lines is perpendicular to the edge of the surface grayscale image; Move along the horizontal direction or vertical direction of the surface grayscale image respectively, and when the straight line passes through the grid line positioning point, the straight line is used as a template grid line; Traverse and slide in sequence until all grid line positioning points are on a straight line, that is, all template grid lines of the solar panel are obtained.

8. The method for identifying grid lines of a solar panel according to claim 1, characterized in that: The template grid line is a single pixel line.

9. A method for identifying grid lines of a solar panel according to claim 1, characterized in that: The pixels whose grayscale values ​​in the window area are greater than the grayscale mean corresponding to the window area are regarded as vacant pixels.

Images