A photovoltaic panel detection method and system based on machine vision

Abstract

The present invention relates to the technical field of target detection, and specifically to a photovoltaic panel detection method and system based on machine vision, which includes the following steps: obtaining photovoltaic panel cell displacement data, calculating the displacement gradient of connection points, dividing rigid and non-rigid regions, generating deformation classification results, constructing a global deformation field, generating a time series deformation trend matrix, calculating the average displacement of bracket fixed points, adjusting deformation compensation parameters, generating adaptive compensation parameters, adjusting the positions of welding points, calculating the Euclidean distance of abnormal regions, and generating defect matching and recognition results. In the present invention, by calculating the displacement gradient of connection points, rigid and non-rigid regions are accurately divided, a global deformation field is constructed by combining the displacement changes of feature points to ensure continuous trends, the compensation parameters are adjusted using the average displacement of fixed points, the deformation increment factor, correction weight, and stability are calculated, the positions of welding points are adjusted to optimize inverse deformation correction, and defects are matched by combining the Euclidean distance, thereby improving the detection accuracy and stability.

Description

Technical Field

[0001] The present invention relates to the technical field of object detection, and particularly to a photovoltaic panel detection method and system based on machine vision. Background Art

[0002] The technical field of object detection includes object recognition, localization, and classification based on computer vision and image processing technologies. The core content of this field is to analyze image or video data to identify specific objects and determine their positions in the image. Object detection technology usually consists of steps such as feature extraction, candidate region generation, and classification. Among them, feature extraction can be based on information such as edges, textures, and colors. Candidate region generation can be achieved through methods such as sliding windows and region generation networks. Classification can be completed using traditional machine learning methods or deep learning models. This technology is widely applied in multiple fields such as autonomous driving, security monitoring, medical image analysis, and industrial inspection, and relies on methods such as deep neural networks, pattern recognition, computer vision, and sensor fusion to continuously develop to improve detection accuracy and computational efficiency.

[0003] Among them, the photovoltaic panel detection method based on machine vision refers to using image analysis and object detection technologies to identify and monitor key components and operating states of electric vehicles. This method mainly monitors multiple technical matters such as battery status, vehicle body integrity, tire wear degree, lighting signals, and dashboard displays. Specific methods include abnormal recognition of battery appearance based on image processing technology, damage detection of vehicle body structure using edge detection and deep learning models, evaluation of tire wear using contour analysis and morphological processing methods, detection of lighting signal status by combining optical flow analysis and color recognition, and parsing of dashboard numerical information through character recognition and template matching technology.

[0004] The prior art is based on static image object detection, does not consider the deformation effects of photovoltaic panels caused by factors such as temperature and mechanical stress, cannot effectively model the dynamic deformation process, and reduces the detection accuracy. The region division relies on edge and texture features, is easily affected by noise interference, and the distinction between rigid and non-rigid regions is not precise enough, affecting the accuracy of deformation compensation. Defect recognition is based on traditional image processing, does not combine deformation information, is prone to misjudging defects due to minor deformations in the glass region, and affects the reliability of recognition. The adjustment of the welding point position ignores the global deformation trend, resulting in the accumulation of compensation errors and affecting the overall correction effect. The deformation correction does not fully combine time series information, resulting in a lack of coherence in the correction results. Under the influence of long-term operation and complex environments, it is difficult to ensure the detection stability and accuracy. Summary of the Invention

[0005] The purpose of the present invention is to solve the disadvantages existing in the prior art, and to propose a photovoltaic panel detection method and system based on machine vision.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A photovoltaic panel detection method based on machine vision, comprising the following steps:

[0007] S1: Obtain the displacement data of multiple cells on the surface of the photovoltaic panel in the video, calculate the displacement vector gradient change rate of the connection points of the photovoltaic cells, divide the rigid region and the non-rigid region, and generate the deformation classification result of the photovoltaic panel region;

[0008] S2: Based on the deformation classification result of the photovoltaic panel region, calculate the displacement change amount of the feature points between adjacent frames of the photovoltaic module encapsulation layer, establish the global deformation field of the photovoltaic panel, and generate the deformation trend matrix of the photovoltaic panel time series;

[0009] S3: Based on the deformation trend matrix of the photovoltaic panel time series, calculate the overall displacement mean value of the fixed points of the photovoltaic bracket, adjust the deformation compensation parameters of the photovoltaic module encapsulation layer, calculate the deformation increment factor of the series connection region of the photovoltaic cells, the correction weight of the non-rigid region of the photovoltaic module, and the deformation stability of the photovoltaic glass surface region, and generate the adaptive deformation compensation parameters of the photovoltaic panel;

[0010] S4: Based on the adaptive deformation compensation parameters of the photovoltaic panel, adjust the positions of the feature points of the welding points of the photovoltaic module, calculate the mapping relationship, and generate the inverse deformation correction image of the photovoltaic panel;

[0011] S5: Based on the inverse deformation correction image of the photovoltaic panel, calculate the Euclidean distance of the morphological features between the abnormal region and the normal region on the surface of the photovoltaic cell, and generate the defect matching and recognition result of the photovoltaic panel.

[0012] The deformation classification result of the photovoltaic panel region includes the rigid region, the non-rigid region, and the displacement vector gradient change rate of the connection points of the photovoltaic cells. The deformation trend matrix of the photovoltaic panel time series includes the displacement change amount of the feature points between adjacent frames of the photovoltaic module encapsulation layer and the global deformation field of the photovoltaic panel. The adaptive deformation compensation parameters of the photovoltaic panel include the overall displacement mean value of the fixed points of the photovoltaic bracket, the deformation compensation parameters of the photovoltaic module encapsulation layer, the deformation increment factor of the series connection region of the photovoltaic cells, the correction weight of the non-rigid region of the photovoltaic module, and the deformation stability of the photovoltaic glass surface region. The inverse deformation correction image of the photovoltaic panel includes the position adjustment of the feature points of the welding points of the photovoltaic module and the mapping relationship. The defect matching and recognition result of the photovoltaic panel includes the Euclidean distance of the morphological features of the abnormal region on the surface of the photovoltaic cell and the Euclidean distance of the morphological features of the normal region on the surface of the photovoltaic cell.

[0013] As a further solution of the present invention, the specific steps for obtaining the deformation classification result of the photovoltaic panel region are as follows:

[0014] S101: Obtain the displacement data of the cells on the surface of the photovoltaic panel, calculate the local displacement vector, obtain the displacement vector gradient of the connection points, and get the displacement vector gradient distribution;

[0015] S102: Call the displacement vector gradient distribution, calculate the gradient value, screen the high and low gradient regions, judge the distribution density, divide the rigid and non-rigid regions, and obtain the regional rigidity classification matrix;

[0016] S103: Call the regional rigidity classification matrix, analyze the connection points in the non-rigid region, calculate the deformation center, combine the rigid region distribution to calculate the deformation classification index, using the formula:

[0017] ;

[0018] Calculate the deformation classification value of the photovoltaic panel region, screen the deformation categories, and obtain the deformation classification result of the photovoltaic panel region;

[0019] Wherein, represents the deformation classification value of the photovoltaic panel region, represents the displacement vector gradient of the th non-rigid region, represents the distance from the region to the adjacent rigid region, is the deformation adjustment parameter, represents the deformation intensity of the th point in the non-rigid region, represents the average deformation intensity of the adjacent rigid regions of the point, represents the total number of non-rigid regions, represents the total number of adjacent rigid regions.

[0020] As a further solution of the present invention, the obtaining steps of the deformation trend matrix of the photovoltaic panel time series are specifically as follows:

[0021] S201: Based on the deformation classification result of the photovoltaic panel region, obtain the feature points between adjacent frames of the encapsulation layer, calculate the displacement change amount of the feature points, and obtain the feature point displacement change matrix;

[0022] S202: Call the feature point displacement change matrix, calculate the deformation vectors of multiple feature points in the global coordinate system, integrate the local deformation information, and establish the global deformation field of the photovoltaic panel;

[0023] S203: Call the global deformation field of the photovoltaic panel, calculate the deformation trend values at multiple time steps in the time series, using the formula:

[0024] ;

[0025] Calculate the deformation characteristics of the time series, and establish the deformation trend matrix of the photovoltaic panel time series;

[0026] Among them, represents the total deformation index under the time series, represents the influence coefficient at the th time step, represents the displacement change at the th time step, represents the displacement deviation correction at the th time step, represents the change sensitivity at the th time step, represents the smoothing parameter of the deformation trend, represents the displacement normalization parameter at the th time step, represents the total number of reference time steps.

[0027] As a further solution of the present invention, the steps for obtaining the adaptive deformation compensation parameters of the photovoltaic panel are specifically as follows:

[0028] S301: Based on the time series deformation trend matrix of the photovoltaic panel, extract the displacement data of the fixed points of the photovoltaic support under the entire time series, and calculate the overall displacement mean value of the photovoltaic support;

[0029] S302: Call the overall displacement mean value of the photovoltaic support, calculate the parameters of the photovoltaic module encapsulation layer, and adjust the parameters according to the compensation requirements of multiple component areas to obtain the deformation compensation parameters of the photovoltaic module;

[0030] S303: Call the deformation compensation parameters of the photovoltaic module and use the formula:

[0031] ;

[0032] Calculate the deformation increment factor of the series connection area of the photovoltaic cells;

[0033] Among them, represents the deformation increment factor of the series connection area of the photovoltaic cells, represents the deformation change value of the th series connection area, represents the deformation weight of the area, is the deformation adjustment parameter, represents the deformation intensity of the th connection point inside the area, represents the mean deformation of the connection points, represents the total number of series connection areas, represents the number of connection points in the area;

[0034] S304: Invoke the deformation increment factor of the series connection area of the photovoltaic cell, calculate the correction weight of the non-rigid area of the photovoltaic module and the deformation stability of the photovoltaic glass surface, and establish the adaptive deformation compensation parameters of the photovoltaic panel.

[0035] As a further solution of the present invention, the steps for obtaining the inverse deformation correction image of the photovoltaic panel are specifically as follows:

[0036] S401: Based on the adaptive deformation compensation parameters of the photovoltaic panel, adjust the position of the characteristic points of the welding points of the photovoltaic module, perform position correction according to the deformation compensation parameters, and obtain the adjustment matrix of the welding points of the photovoltaic module;

[0037] S402: Invoke the adjustment matrix of the welding points of the photovoltaic module, calculate the mapping relationship of the characteristic points before and after adjustment, establish the coordinate transformation relationship for all adjustment points, and use the formula:

[0038] ;

[0039] Calculate the mapping relationship between multiple characteristic points, obtain the spatial mapping matrix after deformation compensation, and obtain the mapping relationship matrix of the photovoltaic module;

[0040] Wherein, represents the mapping relationship value from the th adjusted characteristic point to the th target characteristic point, represents the position offset of the th characteristic point in the st adjustment, represents the correction amount of the th target characteristic point in the st adjustment, is the adjustment offset correction factor, represents the normalization parameter, represents the th adjustment, and represents the position of the th characteristic point in the original coordinates, represents the new position of the same characteristic point after adjustment, represents the th adjusted characteristic point in the th round of transformation, represents the th target characteristic point in the same round of transformation, represents the number of adjustments, represents the total number of characteristic points, represents the total number of transformation operations;

[0041] S403: Call the photovoltaic module mapping relationship matrix, adjust the image pixel coordinates using the transformation parameters, reconstruct the corrected pixel distribution, perform image interpolation to fill the pixel gaps, and generate an inverse deformation corrected image of the photovoltaic panel.

[0042] As a further solution of the present invention, the steps for obtaining the photovoltaic panel defect matching recognition result are specifically as follows:

[0043] S501: Based on the inverse deformation corrected image of the photovoltaic panel, obtain the morphological features of the abnormal area and the normal area on the surface of the photovoltaic cell, extract the key morphological parameters, and obtain the morphological feature set of the photovoltaic cell area;

[0044] S502: Call the morphological feature set of the photovoltaic cell area, calculate the Euclidean distance of the morphological features between the abnormal area and the normal area, perform distance weighted operations for multiple morphological feature dimensions, and use the formula:

[0045] ;

[0046] Calculate the Euclidean distance of the morphological features, screen the feature matching degree of the abnormal area, and obtain the Euclidean distance matrix of the photovoltaic panel morphological features;

[0047] Where, represents the Euclidean distance of the morphological features between the abnormal area and the normal area , represents the weighting factor of the th morphological feature, represents the value of the abnormal area in the th morphological feature dimension, represents the value of the normal area in the same feature dimension, represents the norm parameter for calculating the morphological feature distance, represents the feature matching adjustment parameter, represents the value of the abnormal area in the th auxiliary feature dimension, represents the value of the normal area in the same auxiliary feature dimension, represents the total number of key features, represents the total number of auxiliary features;

[0048] S503: Call the Euclidean distance matrix of the photovoltaic panel morphological features, match the morphological feature relationship between the abnormal area and the normal area, and establish the photovoltaic panel defect matching recognition result.

[0049] A photovoltaic panel detection system based on machine vision, the photovoltaic panel detection system based on machine vision is used to execute the above-mentioned photovoltaic panel detection method based on machine vision, and the system includes:

[0050] The displacement data calculation module obtains the displacement data of the cells on the surface of the photovoltaic panel, calls the connection point coordinate information, calculates the change rate of the displacement vector gradient, screens the points below the threshold to divide the rigid region, screens the points above the threshold to divide the non-rigid region, and integrates the distribution of the rigid and non-rigid regions to obtain the deformation classification result of the photovoltaic panel region;

[0051] The deformation trend analysis module, based on the deformation classification result of the photovoltaic panel region, calls the coordinate of the feature points of adjacent frames of the encapsulation layer, calculates the displacement change amount, and integrates the global deformation field to obtain the deformation trend matrix of the photovoltaic panel time series;

[0052] The deformation compensation calculation module, based on the deformation trend matrix of the photovoltaic panel time series, calls the displacement of the support fixing points, calculates the overall mean value, adjusts the compensation parameters, and calculates the deformation increment factor to obtain the adaptive deformation compensation parameters of the photovoltaic panel;

[0053] The inverse deformation correction module, based on the adaptive deformation compensation parameters of the photovoltaic panel, adjusts the position of the feature points of the welding points, calculates the mapping relationship, and obtains the inverse deformation correction image of the photovoltaic panel;

[0054] The defect recognition module, based on the inverse deformation correction image of the photovoltaic panel, calculates the Euclidean distance between the abnormal region and the normal region, matches the defect feature library, and obtains the defect matching recognition result of the photovoltaic panel.

[0055] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0056] In the present invention, by calculating the change rate of the displacement vector gradient of the connection points of multiple cells of the photovoltaic panel, the rigid and non-rigid regions are finely divided, making the deformation classification more accurate. Combining the displacement changes of the feature points between adjacent frames of the encapsulation layer to establish a global deformation field, making the deformation trend have time continuity. Using the overall displacement mean value of the fixed points to adjust the deformation compensation parameters, making the compensation adapt to the characteristics of different regions. Calculating the deformation increment factor of the series connection region, the correction weight of the non-rigid region, and the deformation stability of the glass surface region, making the compensation more targeted. Adjusting the position of the feature points of the welding points through the mapping relationship to achieve more stable inverse deformation correction. When matching defects, calculating the Euclidean distance of the morphological features between the abnormal and normal regions, making the recognition process combine deformation information and improving the accuracy. The overall solution combines dynamic deformation analysis, global deformation trend modeling, and precise deformation compensation, enabling the photovoltaic panel detection to maintain high accuracy and stability in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a schematic diagram of the working process of the present invention;

[0058] Figure 2 This is the flow chart of the steps for obtaining the deformation classification result of the photovoltaic panel area of the present invention;

[0059] Figure 3 This is the flow chart of the steps for obtaining the deformation trend matrix of the photovoltaic panel time series of the present invention;

[0060] Figure 4 This is the flow chart of the steps for obtaining the adaptive deformation compensation parameters of the photovoltaic panel of the present invention;

[0061] Figure 5 This is the flow chart of the steps for obtaining the inverse deformation correction image of the photovoltaic panel of the present invention;

[0062] Figure 6 This is the flow chart of the steps for obtaining the defect matching recognition result of the photovoltaic panel of the present invention. Detailed implementation manners

[0063] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0064] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation of the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0065] Embodiment 1

[0066] Please refer to Figure 1 , the present invention provides a technical solution: a photovoltaic panel detection method based on machine vision, including the following steps:

[0067] S1: Obtain the displacement data of multiple solar cells on the surface of the photovoltaic panel in the video, calculate the displacement vector gradient change rate of the connection points of the photovoltaic cells, divide the rigid area and the non-rigid area, and generate the deformation classification result of the photovoltaic panel area;

[0068] S2: Based on the deformation classification result of the photovoltaic panel area, calculate the displacement change amount of the feature points between adjacent frames of the photovoltaic module encapsulation layer, establish the global deformation field of the photovoltaic panel, and generate the deformation trend matrix of the photovoltaic panel time series;

[0069] S3: Based on the deformation trend matrix of the photovoltaic panel time series, the overall displacement mean of the photovoltaic bracket fixing point is calculated, the deformation compensation parameters of the photovoltaic module encapsulation layer are adjusted, the deformation increment factor of the photovoltaic cell series connection area, the correction weight of the non-rigid area of ​​the photovoltaic module and the deformation stability of the photovoltaic glass surface area are calculated, and the photovoltaic panel adaptive deformation compensation parameters are generated;

[0070] S4: Based on the photovoltaic panel adaptive deformation compensation parameters, adjust the positions of the characteristic points of the photovoltaic module welding points, calculate the mapping relationship, and generate the photovoltaic panel inverse deformation correction image;

[0071] S5: Based on the inverse deformation correction image of the photovoltaic panel, the Euclidean distance of the morphological features between the abnormal area on the surface of the photovoltaic cell and the normal area on the photovoltaic cell is calculated to generate a photovoltaic panel defect matching recognition result.

[0072] The regional deformation classification results of photovoltaic panels include the displacement vector gradient change rate of rigid areas, non-rigid areas, and photovoltaic cell connection points. The photovoltaic panel time series deformation trend matrix includes the displacement change of feature points between adjacent frames of the photovoltaic module encapsulation layer and the global deformation field of the photovoltaic panel. The adaptive deformation compensation parameters of the photovoltaic panel include the overall displacement mean of the photovoltaic bracket fixing point, the deformation compensation parameters of the photovoltaic module encapsulation layer, the deformation increment factor of the photovoltaic cell series connection area, the correction weight of the non-rigid area of ​​the photovoltaic module, and the deformation stability of the photovoltaic glass surface area. The photovoltaic panel inverse deformation correction image includes the position adjustment and mapping relationship of the feature points of the photovoltaic module welding points. The photovoltaic panel defect matching and recognition results include the Euclidean distance of the morphological features of the abnormal area on the surface of the photovoltaic cell and the Euclidean distance of the morphological features of the normal area of ​​the photovoltaic cell.

[0073] See also Figure 2 , the specific steps for obtaining the regional deformation classification results of photovoltaic panels are:

[0074] S101: Obtain displacement data of the solar cells on the surface of the photovoltaic panel, calculate the local displacement vector, obtain the gradient of the connection point displacement vector, and obtain the gradient distribution of the displacement vector;

[0075] It is necessary to arrange high-precision displacement sensors on different photovoltaic modules. Each sensor is installed on the surface of the cell in a uniform distribution manner and records the displacement of the cell in real time under changes in the external environment (temperature, wind speed, light, etc.). For example, 4 sensors are arranged on the surface of each cell to record the displacement information of its four corner points respectively, and collect data at a time interval of 1s; these data are transmitted to the calculation unit through the data acquisition module, and the calculation unit calculates the data of each monitoring point of each cell to obtain the local displacement vector. For example, for a certain cell, the displacements recorded at its four corner points are mm, the local displacement vector of the cell can be calculated by the average displacement of four points, and its value is mm; Based on the local displacement vector of each cell, calculate the displacement vectors of all adjacent cells, and use the central difference method to obtain the displacement vector gradient of each connection point. Specifically, if the displacement vectors of two adjacent cells are 0.13 mm and 0.09 mm respectively, the gradient calculation method is where is the cell spacing. Assuming mm, the calculated gradient value is mm / mm; Calculate the gradients of all connection points on the photovoltaic panel in the same way to obtain the displacement vector gradient distribution.

[0076] S102: Call the displacement vector gradient distribution, calculate the gradient value, screen the high and low gradient regions, judge the distribution density, divide the rigid and non-rigid regions, and obtain the regional rigidity classification matrix;

[0077] First, calculate the gradient mean value of the entire panel area , assuming that the gradient values of all connection points are: ;

[0078] Then calculate the mean value:

[0079] mm / mm;

[0080] Then compare the gradient value of each connection point with the mean value, and set the judgment threshold . Generally, the threshold is taken as 1.2 times the mean value, that is mm / mm. If the gradient value of a certain connection point is higher than the threshold, it is classified as a high gradient region, otherwise it is classified as a low gradient region. For example, if the connection point gradient values are 0.02, 0.03, 0.05, 0.01, 0.04, then 0.05 and 0.04 are classified as high gradient regions, and the rest are classified as low gradient regions; Then, calculate the distribution density of the high gradient region and the low gradient region. The density calculation method is the number of high gradient points per unit area. Assuming that the area of a certain region is 10 and it contains 3 high gradient points, then the high gradient distribution density of this region is points / cm2; Classify all regions according to the density threshold points / cm2. Regions with density higher than the threshold are classified as non-rigid regions, and regions with density lower than the threshold are classified as rigid regions to obtain the regional rigidity classification matrix.

[0081] S103: Call the regional rigidity classification matrix, analyze the connection points in the non-rigid region, calculate the deformation center, and calculate the deformation classification index in combination with the rigid region distribution. Use the formula:

[0082] ;

[0083] Calculate the deformation classification value of the photovoltaic panel area, screen the deformation categories, and obtain the deformation classification result of the photovoltaic panel area;

[0084] Among them, represents the deformation classification value of the photovoltaic panel area, represents the displacement vector gradient of the th non-rigid area, represents the distance from the area to the adjacent rigid area, is the deformation adjustment parameter, represents the deformation intensity of the th point in the non-rigid area, represents the average deformation intensity of the adjacent rigid area of the point, represents the total number of non-rigid areas, represents the total number of adjacent rigid areas.

[0085] First, calculate the deformation center of the non-rigid area. The deformation center is calculated by the centroid of all connection points in the non-rigid area. Suppose a non-rigid area contains 4 connection points with coordinates (1, 2), (3, 4), (5, 6), and (7, 8) respectively. Then the deformation center is calculated as

[0086] , ;

[0087] That is, the deformation center coordinates are (4, 5); then, calculate the minimum distance from this deformation center to the adjacent rigid area. Suppose the coordinates of the nearest connection point of the adjacent rigid area are (8, 10). Then the distance between the two points is calculated by the Euclidean distance:

[0088] mm;

[0089] Next, calculate the deformation classification index using the formula:

[0090] ;

[0091] Among them, let , the non-rigid area contains 3 connection points with gradients of mm / mm respectively, the distance values are mm respectively, the deformation intensities of the adjacent rigid areas are mm / mm respectively, and the deformation intensities of the non-rigid areas are mm / mm respectively. Then the deformation classification index is calculated as follows:

[0092] 1.

[0093] 2.

[0094] 3.

[0095] 4.

[0096] The deformation classification index value is calculated , if the deformation classification index is greater than the threshold value of 0.75, it is determined that obvious deformation has occurred in this area. Therefore, this area is classified as a deformed area, and finally the deformation classification result of the photovoltaic panel area is obtained.

[0097] Please refer to Figure 3 , the steps for obtaining the deformation trend matrix of the photovoltaic panel time series are specifically as follows:

[0098] S201: Based on the deformation classification result of the photovoltaic panel area, obtain the feature points between adjacent frames of the encapsulation layer, calculate the displacement change amount of the feature points, and obtain the feature point displacement change matrix;

[0099] When the photovoltaic panel operates outdoors for a long time, affected by external environmental changes (temperature, wind force, light intensity, etc.), its encapsulation layer will deform, resulting in displacement differences between the feature points of the photovoltaic module in adjacent time frames. First, determine the positions of several feature points on the surface of the photovoltaic module. Image acquisition can be performed through a high-precision camera device, and the coordinate information of the feature points can be extracted by combining computer vision processing technology. For example, the optical flow method is used to extract the coordinate information of the feature points between time frames. Set the time interval to 0.1 s. If a certain feature point is at the frame with coordinates ( ), and at the frame with coordinates ( ), then the displacement change amount of this feature point is calculated as follows:

[0100] ;

[0101] For example, for a feature point, its coordinates at the frame are (102.3, 205.7), and at the frame are (104.8, 207.2), then calculate its displacement change amount:

[0102] ;

[0103] This calculation needs to be performed batchwise for all feature points on the surface of the photovoltaic module encapsulation layer to form an overall displacement change matrix, as shown in Table 1.

[0104] Table 1 Photovoltaic panel feature point displacement change matrix

[0105]

[0106] As shown in Table 1, the displacement change of each feature point is calculated, and combined with the displacement change matrix of all feature points, the feature point displacement change matrix is obtained.

[0107] S202: Call the feature point displacement change matrix, calculate the deformation vectors of multiple feature points in the global coordinate system, integrate the local deformation information, and establish the global deformation field of the photovoltaic panel;

[0108] Call the feature point displacement change matrix, calculate the deformation vectors of each feature point in the global coordinate system, integrate the local deformation information, and use the two-dimensional coordinate transformation method to project the local displacements of the feature points into the global deformation field. First, set the origin position of the global coordinate system, and use the center point of the photovoltaic module as the global coordinate origin ( ), then, represent the local displacement vector of the feature point as:

[0109] ;

[0110] Among them, , , for example, the coordinate change of a certain feature point is (104.8, 207.2) → (107.3, 209.6), then:

[0111] ;

[0112] Calculate the deformation vector of this feature point:

[0113] ;

[0114] By accumulating and normalizing the deformation vectors of all feature points, a global deformation field is established, as shown in Table 2.

[0115] Table 2 Global deformation field matrix

[0116]

[0117] As shown in Table 2, the local deformation vectors of each feature point have been mapped to the global coordinate system, and finally the global deformation field of the photovoltaic panel is established.

[0118] S203: Call the global deformation field of the photovoltaic panel, calculate the deformation trend values at multiple time steps in the time series, and use the formula:

[0119] ;

[0120] Calculate the time series deformation characteristics and establish the time series deformation trend matrix of the photovoltaic panel;

[0121] Among them, Represents the total deformation index under the time series, Represents the influence coefficient at the th time step, Represents the displacement change at the th time step, Indicates the displacement deviation correction at the th time step, Represents the change sensitivity at the th time step, Indicates the smoothing parameter for the deformation trend, Indicates the displacement normalization parameter at the th time step, Represents the total number of time steps for reference.

[0122] Formula:

[0123] ;

[0124] Calculate the deformation trend index for each time step, set the time interval to 0.1s, assume the deformation monitoring time window for a photovoltaic panel is 10s, that is , given the following data:

[0125] Influence coefficient ;

[0126] Displacement change ;

[0127] Deviation correction ;

[0128] Change sensitivity ;

[0129] Adjustment parameter ;

[0130] Normalization parameter ;

[0131] Substitute into the calculation:

[0132] ;

[0133] Calculate the numerator part:

[0134] ;

[0135] Calculate the denominator part:

[0136] ;

[0137] Final calculation:

[0138] ;

[0139] The result shows that under the current time series window, the deformation trend index of the photovoltaic panel is 4.12. This value can be used to monitor the overall deformation degree of the photovoltaic panel during operation. Combining historical data analysis can further determine whether the deformation trend is within the normal range, and finally establish a time series deformation trend matrix for the photovoltaic panel.

[0140] Please refer to Figure 4 , and the specific steps for obtaining the adaptive deformation compensation parameters of the photovoltaic panel are as follows:

[0141] S301: Based on the time series deformation trend matrix of the photovoltaic panel, extract the displacement data of the fixed points of the photovoltaic support under the entire time series, and calculate the overall displacement mean value of the photovoltaic support;

[0142] First, extract the displacement data of the fixed points of the photovoltaic support in the entire time series. During the extraction process, by setting multiple fixed monitoring points, each monitoring point needs to record the displacement change amount at a series of time points. The time interval is set to 1 hour, and 24 times are recorded daily, and the data for 30 days are accumulated to form a long-term trend matrix. For the displacement data of each monitoring point, it is necessary to calculate its displacement increment respectively, that is, the difference between the displacement value at the current time point and the displacement value at the previous time point, to form a time series displacement change curve. When calculating the overall displacement mean value of the photovoltaic support, first sum up the displacement data of all fixed points, and then divide by the total number of fixed points to obtain the single-day mean value, and further calculate the mean trend for 30 days. This mean trend is calculated using the moving average method, and the window length is set to 7 days to reduce the influence of short-term data fluctuations. The selection of this window length is determined based on the historical monitoring data fluctuation range, that is, if the standard deviation of the 7-day moving average is less than 0.02 mm, the 7-day window is maintained, otherwise it is expanded to 10 days. If the displacement increment of a certain fixed point exceeds 1.5 times the overall mean value, it is marked as an abnormal point, and the validity of its data source is re-evaluated. The setting of this 1.5-fold threshold is determined based on the deformation limit of the photovoltaic support material. For example, the allowable elastic deformation of an aluminum alloy support generally does not exceed 0.3 mm. Therefore, by calculating the maximum fluctuation range of the support under no external force through historical data, the average fluctuation upper limit is obtained to be about 0.2 mm, and after taking 1.5 times, it is set to 0.3 mm to ensure that the threshold can identify abnormal data without misjudging normal fluctuations. Calculate the overall displacement mean value of the photovoltaic support through the screened data, as shown in Table 3.

[0143] Table 3 Displacement monitoring data of the fixed points of the photovoltaic support

[0144]

[0145] As shown in Table 3, through the monitoring of multi-day data, it can be found that the overall displacement mean value of the photovoltaic support is about 0.13 mm. This data will be used for subsequent calculation of the adjustment of the photovoltaic module encapsulation layer parameters.

[0146] S302: Call the overall displacement mean value of the photovoltaic support, calculate the parameters of the photovoltaic module encapsulation layer, adjust the parameters according to the compensation requirements of multiple component areas, and obtain the deformation compensation parameters of the photovoltaic module;

[0147] Call the overall displacement mean value of the photovoltaic support and calculate based on the compensation requirements for component area deformation. First, for the encapsulation layer of the photovoltaic module, the deformation amount of the component area needs to be obtained. Set the measurement interval to 1 hour, record the surface displacement amount of each component area, calculate the mean value of the component area deformation through matrix calculation, and perform a difference calculation with the overall displacement mean value of the photovoltaic support to obtain the deformation adjustment value of the component encapsulation layer. If the deformation amount of a certain component area exceeds 20% of the overall mean value, the compensation weight needs to be increased. This 20% threshold is set based on the allowable stress distribution range of the photovoltaic encapsulation layer. According to material mechanics calculations, the thermal expansion rate of the photovoltaic encapsulation layer under the influence of environmental temperature difference is about 0.04%. Considering a daily temperature difference of 30°C, the maximum deformation change rate is about 1.2%. Therefore, 20% is the optimized value after empirical adjustment. This weight is calculated using the proportional correction method, and the weight setting range is 1.0 - 2.5, with the reference value set to 1.5. This reference value is set based on the measurement of the impact of component area deformation on power generation efficiency without compensation. If the deformation amount is greater than 30% of the overall mean value, the component power attenuation is about 2.3%. Therefore, the reference value of 1.5 can moderately reduce the deformation impact. If the deformation amount is less than 5% of the mean value, the weight is set to 1.0. If the deformation amount exceeds 30% of the mean value, it is set to 2.5. When calculating the deformation compensation parameters of the photovoltaic module encapsulation layer, the final compensation parameters are calculated based on the deformation adjustment values and weights of each area, as shown in Table 4.

[0148] Table 4 Calculation of Deformation Compensation Parameters of Photovoltaic Module Encapsulation Layer

[0149]

[0150] S303: Call the deformation compensation parameters of the photovoltaic module and use the formula:

[0151] ;

[0152] Calculate the deformation increment factor of the series connection area of the photovoltaic cells;

[0153] where, represents the deformation increment factor of the series connection area of the photovoltaic cells, represents the deformation change value of the th series connection area, represents the deformation weight of the area, is the deformation adjustment parameter, represents the deformation intensity of the th connection point inside the area represents the average deformation of the connection points, represents the total number of series connection regions, represents the number of connection points in the region;

[0154] Formula:

[0155] ;

[0156] In this calculation process, first, the deformation change value of the series connection region needs to be determined , which is derived from the displacement change of the photovoltaic module region and obtained by using the differential calculation method, that is:

[0157] ;

[0158] where and are the deformation values at the current moment and the previous moment respectively, and the deformation weight is set according to the regional influence degree, and the value range is 0.8 - 2.0. The adjustment parameter is set according to the regional connection strength, and the value range is 0.5 - 1.5. The deformation strength of the connection points is obtained from the monitoring data, and the average deformation is calculated by taking the average of all connection point deformation values;

[0159] In the calculation process of the deformation increment factor of the series region of photovoltaic cells, first determine the deformation change value , which is determined by the difference in deformation between the current moment and the previous moment using the differential calculation method. In the example, take , , and get . The deformation weight is set according to the regional influence degree, and the range is 0.8 - 2.0. In the example, take . The adjustment parameter reflects the regional connection strength, and the range is 0.5 - 1.5. In the example, take . The deformation strength of the connection points is obtained from the monitoring data, and the average deformation is calculated from all connection point deformation values. In the example, take , .

[0160] Substitute the parameters:

[0161] ;

[0162] ;

[0163] ;

[0164] The deformation increment factor of the series connection area of the photovoltaic cell is calculated .

[0165] S304: Invoke the deformation increment factor of the series connection area of the photovoltaic cell, calculate the correction weight of the non-rigid area of the photovoltaic module and the deformation stability of the photovoltaic glass surface, and establish the adaptive deformation compensation parameters of the photovoltaic panel.

[0166] The correction weight of the non-rigid area is set according to the magnitude of the deformation increment factor. The setting of this correction weight refers to the influence of the power attenuation of the module under different deformations. If , the correction weight is set to 0.9. If , the correction weight is set to 1.0. If , the correction weight is set to 1.2. At the same time, calculate the deformation stability of the photovoltaic glass surface. This stability calculation is based on the standard deviation analysis of the deformation mean value in 30 days. If the standard deviation of the daily deformation amount is less than 0.02 mm, the deformation stability is set to high. If the standard deviation is between 0.02 - 0.05 mm, the deformation stability is set to medium. If the standard deviation is greater than 0.05 mm, the deformation stability is set to low. In this calculation, the deformation standard deviation is 0.03 mm, so the deformation stability is set to medium. Finally, based on the correction weights of each area, establish the adaptive deformation compensation parameters of the photovoltaic panel.

[0167] Please refer to Figure 5 , the specific steps for obtaining the inverse deformation correction image of the photovoltaic panel are as follows:

[0168] S401: Based on the adaptive deformation compensation parameters of the photovoltaic panel, adjust the characteristic point positions of the welding points of the photovoltaic module, perform position correction according to the deformation compensation parameters, and obtain the adjustment matrix of the welding points of the photovoltaic module;

[0169] For each welding point characteristic point, use image analysis technology to identify its pixel position and convert it to the actual physical coordinate system. Assume that the initial coordinates of a certain welding point are (120.5 mm, 80.3 mm). Combining with the adaptive deformation compensation parameters of the photovoltaic panel, the deformation compensation vector of this point can be expressed as (Δx = 2.1 mm, Δy = -1.5 mm). Thus, the corrected coordinates of this point can be calculated as (122.6 mm, 78.8 mm). Perform the same calculation for all welding point characteristic points and record the position information before and after compensation in a table, as shown in Table 5.

[0170] Table 5 Data table for position adjustment of welding point characteristic points

[0171]

[0172] As shown in Table 5, the coordinate of each feature point is corrected after deformation compensation. The corrected coordinate data of all welding points are called to construct the adjustment matrix of the welding points of the photovoltaic module.

[0173] S402: Call the adjustment matrix of the welding points of the photovoltaic module, calculate the mapping relationship of the feature points before and after adjustment, establish the coordinate transformation relationship for all adjustment points, and use the formula:

[0174] ;

[0175] Calculate the mapping relationship between multiple feature points, obtain the spatial mapping matrix after deformation compensation, and get the mapping relationship matrix of the photovoltaic module;

[0176] Among them, represents the mapping relationship value from the th adjusted feature point to the th target feature point, represents the position offset of the th feature point in the th adjustment, represents the correction amount of the th target feature point in the th adjustment, is the adjustment offset correction factor, represents the normalization parameter, represents the position of the th feature point in the original coordinates at the th adjustment, represents the new position of the same feature point after adjustment, represents the mapping smoothing parameter, represents the coordinate value of the th adjusted feature point in the th round of transformation, represents the coordinate value of the th target feature point in the same round of transformation, represents the number of adjustments, represents the total number of feature points, represents the total number of transformation operations;

[0177] Call the adjustment matrix of the welding points of the photovoltaic module, calculate the mapping relationship of the feature points before and after adjustment, and select the set of welding point feature points {P1, P2,..., Pn}. Among them, the coordinates of each point before and after adjustment are and , respectively. When calculating the mapping relationship, the deformation compensation vector and the spatial mapping matrix of the welding points need to be considered, and the formula is used:

[0178] ;

[0179] Calculate the mapping relationship between multiple feature points, obtain the spatial mapping matrix after deformation compensation, and get the photovoltaic module mapping relationship matrix.

[0180] Assign values to the parameters in the formula for calculation:

[0181] Let mm, mm;

[0182] Take , , , ;

[0183] Let , , ;

[0184] Calculate:

[0185]

[0186]

[0187]

[0188]

[0189]

[0190]

[0191]

[0192]

[0193]

[0194]

[0195]

[0196]

[0197]

[0198] Calculate an element of the photovoltaic module mapping relationship matrix , and this result shows that the mapping relationship of the welding points is relatively tight after adjustment, and the spatial mapping matrix after deformation compensation can accurately describe the position relationship of the adjusted feature points.

[0199] S403: Call the photovoltaic module mapping relationship matrix, apply transformation parameters to adjust the image pixel coordinates, reconstruct the corrected pixel distribution, perform image interpolation to fill the pixel gaps, and generate the inverse deformation correction image of the photovoltaic panel.

[0200] First, it is necessary to obtain the projection area of the photovoltaic module in the image. This projection area is determined by factors such as the geometric shape, installation angle, and illumination direction of the photovoltaic module. Specifically, it is necessary to extract the actual position information of the photovoltaic module in the three-dimensional coordinate system and calculate its projection boundary in the image coordinate system based on the optical parameters of the sensor (including focal length, principal point coordinates, distortion coefficients, etc.). Subsequently, based on the obtained projection area of the photovoltaic module, a one-to-one correspondence is established with the pixel coordinates in the image. This process involves coordinate transformation. The spatial coordinates of the photovoltaic module are represented by homogeneous coordinates, and they are subjected to perspective projection transformation through the external parameter matrix and the internal parameter matrix to calculate the projection points in the image coordinates. If the photovoltaic module has an inclination angle, projection distortion needs to be further considered and corrected using affine transformation. The transformation matrix involved in this process can be corrected from measured data. For example, by selecting multiple feature points in the image and combining the actual installation data of the photovoltaic module, the transformation matrix is optimized using the least squares method. The specific optimization process is as follows: 1) Select at least 4 known matching point pairs. 2) Construct an error function, which represents the squared error between the transformed points and the actual points. 3) Calculate the partial derivatives of the error with respect to each parameter of the transformation matrix. 4) Use gradient descent to iteratively update the parameters until convergence within the threshold ε = 0.001 to obtain the final mapping relationship matrix. After completing the adjustment of the image coordinates, reconstruct the corrected pixel distribution. This step needs to ensure the integrity of the pixels, that is, all adjusted pixel points still need to be within the valid area of the image, and at the same time, it is necessary to ensure that the pixel gray values are within a reasonable range (0 - 255). If the pixels overlap, the average value of the overlapping pixels is calculated for fusion. If some pixel points are missing due to the transformation, image interpolation needs to be performed to fill the pixel gaps. Here, the bilinear interpolation method is used, that is, for the missing pixel points, their values are calculated by weighted averaging the gray values of the four surrounding known pixel points. The calculation formula is:

[0201] ;

[0202] where, is the gray value of the missing pixel point, are the gray values of the four surrounding known pixel points respectively, and are the normalized position parameters of the pixel in the horizontal and vertical directions. For example, if a missing pixel point is at the coordinate (12.3, 45.7), and its four adjacent known pixel points are (12, 45), (13, 45), (12, 46), (13, 46) respectively, then , substitute into the formula to calculate the gray value of the missing point, and generate the photovoltaic panel inverse deformation correction image, which is used for subsequent analysis and processing.

[0203] See also Figure 6 , the specific steps for obtaining the photovoltaic panel defect matching and identification results are as follows:

[0204] S501: based on the photovoltaic panel inverse deformation correction image, the morphological features of the abnormal area and the normal area on the surface of the photovoltaic cell are obtained, key morphological parameters are extracted, and a set of morphological features of the photovoltaic cell area is obtained;

[0205] First, several key areas in the image are selected as reference points. The selection of reference points can be based on parameters such as the uniformity of the cell surface texture, the consistency of color distribution, and the characteristics of light reflection. For example, among 100 photovoltaic cell samples, there may be 10-15 abnormal areas, and the brightness value distribution of these areas is different. When extracting features, it is necessary to quantify the edge contour, grayscale gradient change, and area morphology parameters of each area, and set the edge sharpness threshold of the abnormal area to between 50 and 100. When the sharpness value is higher than this range, there may be cracks or foreign objects blocking the edge. The threshold is set based on the surface detection data of standard photovoltaic modules. Specifically, By measuring the sharpness distribution of 500 normal photovoltaic cells and 500 abnormal cells, the normal range is between 40-90, while the abnormal area is usually between 55-120, so 50-100 is selected as the detection range. By calculating the morphological parameters such as the major axis, minor axis ratio, eccentricity and boundary connectivity of each area, its classification can be preliminarily judged, and then statistical normalization processing is performed to facilitate further matching analysis in subsequent steps. The normalization range of the morphological parameters is set to [0,1]. The normalization calculation adopts the minimum-maximum standardization method, and the normalization reference value is set to the range of the minimum and maximum values ​​of the normal photovoltaic module characteristics. That is, the normalization formula is:

[0206] ;

[0207] For example, if the area parameter of an abnormal area is 1200px2, and the area distribution range of the normal area is 800-2000px2, the normalized value is:

[0208] ;

[0209] All calculated morphological characteristic parameters are summarized to obtain the photovoltaic cell regional morphological characteristic set.

[0210] S502: Calling the photovoltaic cell area morphological feature set, calculating the Euclidean distance of the morphological features between the abnormal area and the normal area, and performing distance weighted operations on multiple morphological feature dimensions, using the formula:

[0211] ;

[0212] Calculate the Euclidean distance of morphological features, screen the feature matching degree of abnormal regions, and obtain the Euclidean distance matrix of the morphological features of the photovoltaic panel;

[0213] Among them, represents the Euclidean distance of morphological features between the abnormal region and the normal region , represents the weighting factor of the th morphological feature, represents the value of the abnormal region on the th morphological feature dimension, represents the value of the normal region on the same feature dimension, represents the norm parameter for calculating the morphological feature distance, represents the feature matching adjustment parameter, represents the value of the abnormal region on the th auxiliary feature dimension, represents the value of the normal region on the same auxiliary feature dimension, represents the total number of key features, represents the total number of auxiliary features;

[0214] In the process of calculating the morphological features of photovoltaic cells, it is necessary to calculate the Euclidean distance of morphological features between the abnormal region and the normal region. To achieve this calculation, first, it is necessary to obtain the morphological feature datasets of the abnormal region and the normal region. The morphological feature data of the abnormal region can be extracted by image processing methods. For example, after using an edge detection algorithm to determine the region contour, morphological methods are used to calculate basic morphological parameters such as area, perimeter, and aspect ratio. The morphological feature data of the normal region can be extracted from a large number of normal samples through statistical methods and a reference database is established. After the data extraction is completed, it is necessary to match the morphological feature data of the abnormal region and the normal region for subsequent calculation of their Euclidean distance. In the calculation process, the difference of each morphological feature dimension needs to be weighted. Set the weighting factor of each morphological feature. This weight value can be set according to the importance of the feature for defect recognition. For example, area and aspect ratio may have higher discrimination ability, so a larger weight is given, while secondary features such as edge complexity can be set with a lower weight. For the norm parameter of the morphological feature distance calculation, its value determines the calculation method. For example, corresponds to the standard Euclidean distance, while corresponds to the Manhattan distance. Parameter The setting of will affect the sensitivity of the abnormal area matching the normal area. After calculating the morphological feature distance, it is also necessary to consider the matching of auxiliary features. For example, color, texture, etc. can be used as auxiliary features to improve the matching accuracy. These auxiliary features are calculated by the Euclidean distance and use the parameter to control its influence weight on the overall matching. Suppose an abnormal area has morphological feature data of , and the normal area has morphological feature data of . The corresponding weight factor is set as and . The auxiliary feature weight is set for calculation. Then, according to the following formula:

[0215] ;

[0216] Suppose the specific values are as follows:

[0217] Table 6 Morphological Feature Data Table

[0218]

[0219] As shown in Table 6, the values of different morphological features in the abnormal area and the normal area and their corresponding weights are listed. The weight factor is used to adjust the influence of each feature on the matching calculation.

[0220] Table 7 Auxiliary Feature Data Table

[0221]

[0222] As shown in Table 7, the values of the auxiliary features in the abnormal area and the normal area are listed. These features will be considered when calculating the Euclidean distance.

[0223] Calculate the morphological feature distance:

[0224] ;

[0225] ;

[0226] ;

[0227] Calculate the auxiliary feature distance:

[0228] ;

[0229] ;

[0230] Final Euclidean distance:

[0231] ;

[0232] The result shows that the Euclidean distance of the morphological features between the abnormal area and the normal area is 0.637. This distance value can be used to further classify whether they belong to the same type of area.

[0233] S503: Invoke the Euclidean distance matrix of the morphological features of the photovoltaic panel, match the morphological feature relationship between the abnormal area and the normal area, and establish the matching recognition result of the photovoltaic panel defect.

[0234] Invoke the Euclidean distance matrix of the morphological features of the photovoltaic panel, match the morphological feature relationship between the abnormal area and the normal area. For matching, it is necessary to sort according to the data in the Euclidean distance matrix and set a matching threshold , which is set to 0.08. The setting basis is based on the matching error analysis of 5000 groups of photovoltaic cell defect samples. Under different threshold settings, the error rate curve shows that when takes 0.08, the mean value of the false positive rate and the false negative rate is the lowest. Assume that a certain photovoltaic cell has 5 abnormal areas, which are respectively matched with 10 normal areas, and the calculated Euclidean distances are shown in Table 8:

[0235] Table 8 Morphological Feature Euclidean Distance Matching Table

[0236]

[0237] As shown in Table 8, according to the calculation results of the morphological feature Euclidean distance, the matching situations between the abnormal area and the normal area are as follows: Area A matches normal areas 1 and 7 (because its Euclidean distance is lower), Area B matches normal areas 3 and 5, Area C matches normal areas 4 and 5, Area D matches normal areas 2, 3, and 7, and Area E matches normal areas 4, 5, and 6.

[0238] Finally, through the calculation of the morphological feature Euclidean distance, establish the matching relationship between the photovoltaic panel defect area and the normal area, providing a reference for subsequent defect recognition.

[0239] A photovoltaic panel detection system based on machine vision, the photovoltaic panel detection system based on machine vision is used to execute the above-mentioned photovoltaic panel detection method based on machine vision. The system includes:

[0240] The displacement data calculation module obtains the displacement data of the cells on the surface of the photovoltaic panel, invokes the connection point coordinate information, calculates the displacement vector gradient change rate, screens the points below the threshold to divide the rigid area, screens the points above the threshold to divide the non-rigid area, and integrates the distribution of the rigid and non-rigid areas to obtain the classification result of the photovoltaic panel area deformation;

[0241] Based on the deformation classification results of the photovoltaic panel area, the deformation trend analysis module calls the coordinate of feature points of adjacent frames in the encapsulation layer, calculates the displacement change amount, integrates the global deformation field, and obtains the deformation trend matrix of the photovoltaic panel time series;

[0242] Based on the deformation trend matrix of the photovoltaic panel time series, the deformation compensation calculation module calls the displacement of the support fixing point, calculates the overall average value, adjusts the compensation parameters, calculates the deformation increment factor, and obtains the adaptive deformation compensation parameters of the photovoltaic panel;

[0243] Based on the adaptive deformation compensation parameters of the photovoltaic panel, the inverse deformation correction module adjusts the position of the feature points of the welding points, calculates the mapping relationship, and obtains the inverse deformation correction image of the photovoltaic panel;

[0244] Based on the inverse deformation correction image of the photovoltaic panel, the defect recognition module calculates the Euclidean distance between the abnormal area and the normal area, matches the defect feature library, and obtains the defect matching recognition result of the photovoltaic panel.

[0245] The above is only the preferred embodiment of the present invention, and does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A photovoltaic panel detection method based on machine vision, characterized in that: The following steps are involved: S1: Obtain the displacement data of multiple cells on the surface of the photovoltaic panel in the video, calculate the displacement vector gradient change rate of the photovoltaic cell connection point, divide the rigid area and the non-rigid area, and generate the photovoltaic panel area deformation classification result; The steps for obtaining the photovoltaic panel regional deformation classification result are specifically as follows: S101: Obtain displacement data of the solar cells on the surface of the photovoltaic panel, calculate the local displacement vector, obtain the gradient of the connection point displacement vector, and obtain the gradient distribution of the displacement vector; S102: calling the displacement vector gradient distribution, calculating the gradient value, screening high and low gradient areas, determining the distribution density, dividing the rigid and non-rigid areas, and obtaining the regional rigidity classification matrix; S103: calling the regional rigid classification matrix, analyzing the non-rigid regional connection points, calculating the deformation center, and calculating the deformation classification index in combination with the rigid regional distribution, using the formula: ； Calculate the regional deformation classification value of the photovoltaic panel, filter the deformation category, and obtain the regional deformation classification result of the photovoltaic panel; in, Represents the regional deformation classification value of the photovoltaic panel, Representative The displacement vector gradient of the non-rigid region, represents the distance from the region to the adjacent rigid region, Adjust the parameters for the deformation. Represents the first The deformation strength of a point, represents the average deformation strength of the rigid region adjacent to the point, represents the total number of non-rigid regions, represents the total number of adjacent rigid regions; S2: Based on the regional deformation classification result of the photovoltaic panel, the displacement change of the feature points between adjacent frames of the photovoltaic module encapsulation layer is calculated, the global deformation field of the photovoltaic panel is established, and the time series deformation trend matrix of the photovoltaic panel is generated; S3: Based on the photovoltaic panel time series deformation trend matrix, the overall displacement mean of the photovoltaic bracket fixing point is calculated, the deformation compensation parameters of the photovoltaic module encapsulation layer are adjusted, the deformation increment factor of the photovoltaic cell series connection area, the correction weight of the photovoltaic module non-rigid area and the deformation stability of the photovoltaic glass surface area are calculated, and the photovoltaic panel adaptive deformation compensation parameters are generated; S4: Based on the photovoltaic panel adaptive deformation compensation parameters, adjusting the positions of the characteristic points of the photovoltaic module welding points, calculating the mapping relationship, and generating a photovoltaic panel inverse deformation correction image; S5: Based on the photovoltaic panel inverse deformation correction image, the Euclidean distance of the morphological features between the abnormal area on the photovoltaic cell surface and the normal area on the photovoltaic cell surface is calculated to generate a photovoltaic panel defect matching recognition result.

2. The photovoltaic panel detection method based on machine vision according to claim 1, characterized in that: The photovoltaic panel regional deformation classification results include the displacement vector gradient change rate of the rigid area, non-rigid area, and photovoltaic cell connection point; the photovoltaic panel time series deformation trend matrix includes the displacement change of the feature points between adjacent frames of the photovoltaic module encapsulation layer and the global deformation field of the photovoltaic panel; the photovoltaic panel adaptive deformation compensation parameters include the overall displacement mean of the photovoltaic bracket fixing point, the deformation compensation parameters of the photovoltaic module encapsulation layer, the deformation increment factor of the photovoltaic cell series connection area, the correction weight of the photovoltaic module non-rigid area, and the deformation stability of the photovoltaic glass surface area; the photovoltaic panel inverse deformation correction image includes the position adjustment and mapping relationship of the feature points of the photovoltaic module welding point; the photovoltaic panel defect matching and identification results include the Euclidean distance of the morphological features of the abnormal area on the surface of the photovoltaic cell and the Euclidean distance of the morphological features of the normal area of ​​the photovoltaic cell.

3. The photovoltaic panel detection method based on machine vision according to claim 2, characterized in that: The specific steps for obtaining the photovoltaic panel time series deformation trend matrix are as follows: S201: based on the photovoltaic panel regional deformation classification result, obtaining feature points between adjacent frames of the encapsulation layer, calculating the displacement change of the feature points, and obtaining a feature point displacement change matrix; S202: calling the feature point displacement change matrix, calculating the deformation vectors of multiple feature points in the global coordinate system, integrating local deformation information, and establishing a global deformation field of the photovoltaic panel; S203: calling the photovoltaic panel global deformation field, calculating the deformation trend values ​​of multiple time steps in the time series, using the formula: ； Calculate the time series deformation characteristics and establish the photovoltaic panel time series deformation trend matrix; in, Represents the total deformation index under the time series, Representative The influence coefficient of the next time step, Representative The displacement change in the time step is, Indicates The displacement deviation correction amount of the time step, Representative The sensitivity of the change in the time step, Smoothing parameters that represent deformation trends, Indicates The displacement normalization parameter, Represents the total number of time steps for the reference.

4. The photovoltaic panel detection method based on machine vision according to claim 3, characterized in that: The steps for obtaining the photovoltaic panel adaptive deformation compensation parameters are specifically as follows: S301: Based on the photovoltaic panel time series deformation trend matrix, extract the displacement data of the photovoltaic bracket fixing point in the entire time series, and calculate the overall displacement mean of the photovoltaic bracket; S302: calling the overall displacement mean of the photovoltaic support, calculating the parameters of the photovoltaic module encapsulation layer, adjusting the parameters according to the compensation requirements of multiple module areas, and obtaining the photovoltaic module deformation compensation parameters; S303: Calling the photovoltaic module deformation compensation parameter, using the formula: ； Calculate the deformation increment factor of the photovoltaic cell series area; in, Represents the deformation increment factor of the photovoltaic cell series connection area, Representative The deformation change value of the series-connected area, represents the deformation weight of the region, Adjust the parameters for the deformation. Representing the region The deformation strength of the connection point, represents the mean deformation of the connection point, represents the total number of serially connected regions, The number of connection points representing the region; S304: calling the deformation increment factor of the photovoltaic cell series area, calculating the correction weight of the non-rigid area of ​​the photovoltaic module and the deformation stability of the photovoltaic glass surface, and establishing the photovoltaic panel adaptive deformation compensation parameters.

5. The photovoltaic panel detection method based on machine vision according to claim 4, characterized in that: The steps for obtaining the photovoltaic panel inverse deformation correction image are specifically as follows: S401: adjusting the position of the characteristic point of the photovoltaic module welding point based on the photovoltaic panel adaptive deformation compensation parameter, performing position correction according to the deformation compensation parameter, and obtaining the photovoltaic module welding point adjustment matrix; S402: calling the photovoltaic module welding point adjustment matrix, calculating the mapping relationship of the feature points before and after the adjustment, and establishing a coordinate transformation relationship for all adjustment points, using the formula: ； Calculate the mapping relationship between multiple feature points, obtain the spatial mapping matrix after deformation compensation, and obtain the photovoltaic module mapping relationship matrix; in, Representative Adjust the feature points to The mapping relationship value of the target feature points, Representative The feature point is The position offset in this adjustment, Representative The target feature point is The amount of correction in this adjustment, To adjust the offset correction factor, represents the normalization parameter, Representative The first adjustment The position of feature points in the original coordinates, Represents the new position of the same feature point after adjustment, represents the mapping smoothing parameter, Representative The adjusted feature points are The coordinate values ​​in the round transformation, Representative The coordinate values ​​of the target feature points in the same round of transformation, Represents the number of adjustments, Represents the total number of feature points, Represents the total number of transformation operations; S403: calling the photovoltaic assembly mapping relationship matrix, applying transformation parameters to adjust the image pixel coordinates, reconstructing the corrected pixel distribution, performing image interpolation to fill the pixel gaps, and generating an inverse deformation correction image of the photovoltaic panel.

6. The photovoltaic panel detection method based on machine vision according to claim 5, characterized in that: The steps for obtaining the photovoltaic panel defect matching and identification results are specifically as follows: S501: based on the photovoltaic panel inverse deformation correction image, obtaining morphological features of abnormal areas and normal areas on the surface of the photovoltaic cell, extracting key morphological parameters, and obtaining a photovoltaic cell area morphological feature set; S502: calling the photovoltaic cell area morphological feature set, calculating the Euclidean distance of the morphological features between the abnormal area and the normal area, and performing distance weighted operations on multiple morphological feature dimensions, using the formula: ； Calculate the Euclidean distance of morphological features, screen the feature matching degree of abnormal areas, and obtain the Euclidean distance matrix of photovoltaic panel morphological features; in, Represents abnormal area With normal area The Euclidean distance of the morphological features between Representative The weighting factor of the morphological features, Represents abnormal area In the The value of the morphological feature dimension, Represents the normal area The values ​​on the same feature dimension, represents the norm parameter of the morphological feature distance calculation, represents the feature matching adjustment parameter, Represents abnormal area In the The values ​​on the auxiliary feature dimensions, Represents the normal area The values ​​on the same auxiliary feature dimension, Represents the total number of key features, Represents the total number of auxiliary features; S503: calling the Euclidean distance matrix of the photovoltaic panel morphological features, matching the morphological feature relationship between the abnormal area and the normal area, and establishing a photovoltaic panel defect matching and recognition result.

7. A photovoltaic panel detection system based on machine vision, characterized in that: According to the photovoltaic panel detection method based on machine vision according to any one of claims 1 to 6, the system comprises: The displacement data calculation module obtains the displacement data of the solar cells on the surface of the photovoltaic panel, calls the connection point coordinate information, calculates the displacement vector gradient change rate, selects points below the threshold to divide the rigid area, selects points above the threshold to divide the non-rigid area, integrates the rigid and non-rigid area distribution, and obtains the photovoltaic panel area deformation classification result; The deformation trend analysis module, based on the deformation classification result of the photovoltaic panel area, calls the coordinates of the feature points of the adjacent frames of the encapsulation layer, calculates the displacement change, integrates the global deformation field, and obtains the time series deformation trend matrix of the photovoltaic panel; The deformation compensation calculation module calls the displacement of the bracket fixed point based on the photovoltaic panel time series deformation trend matrix, calculates the overall mean, adjusts the compensation parameters, calculates the deformation increment factor, and obtains the photovoltaic panel adaptive deformation compensation parameters; The inverse deformation correction module adjusts the position of the characteristic points of the welding points based on the adaptive deformation compensation parameters of the photovoltaic panel, calculates the mapping relationship, and obtains the inverse deformation correction image of the photovoltaic panel; The defect recognition module calculates the Euclidean distance between the abnormal area and the normal area based on the inverse deformation correction image of the photovoltaic panel, matches the defect feature library, and obtains the photovoltaic panel defect matching recognition result.

Images