Solar cell panel defect detection method and system based on structured light imaging

Abstract

The invention provides a solar cell panel defect detection method and device for structured light imaging, and relates to the technical field of solar cell panels. According to the method, after a solar cell panel is irradiated through a multi-angle light source, a camera at a fixed position is used for obtaining a structured light image corresponding to the solar cell panel; and photoluminescence difference data corresponding to the solar cell panel is accurately obtained by using the structured light image, so that a high-precision defect detection process in a full scene is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of solar cell panels, and in particular to a method and system for detecting defects in solar cell panels. Background Art

[0002] The existing invisible defect detection process of solar panels is mainly achieved by electroluminescence imaging and photoluminescence imaging. For electroluminescence imaging, the detection process requires applying voltage to the solar panel, which depends to a certain extent on the metallization degree and manufacturing process of the solar panel; and current injection will lead to uneven distribution of carrier concentration, affecting the luminescence intensity, thereby reducing image quality. In addition, the detection process requires direct contact with the solar panel, increasing the risk of component damage or contamination. For photoluminescence imaging, its detection effect is better under indoor or darkroom conditions, but because photovoltaic panels are mostly used in outdoor environments, defect detection under high irradiance conditions is difficult to achieve. Summary of the invention

[0003] In view of this, an object of the present invention is to provide a method and system for solar panel defect detection using structured light imaging. The method uses a camera at a fixed position to obtain a structured light image corresponding to the solar panel after irradiating the solar panel with light sources from multiple angles, and uses the structured light image to accurately obtain photoluminescence difference data corresponding to the solar panel, thereby realizing a high-precision defect detection process in all scenarios.

[0004] In a first aspect, an embodiment of the present invention provides a method for detecting defects in a solar panel using structured light imaging, the method comprising:

[0005] Obtain a solar panel at a preset fixed position, determine a camera and a light source corresponding to the solar panel, and determine a shooting position of the camera according to the fixed position;

[0006] Determine fixed parameters of the solar panel and position parameters of the light source based on the shooting position, and use the fixed parameters and position parameters to control the camera to shoot the solar panel to obtain a structured light image corresponding to the solar panel;

[0007] Extracting a shared feature region contained in the structured light image according to material parameters and lighting parameters of the solar panel, and determining photoluminescence difference data corresponding to the structured light image using the shared feature region;

[0008] Defect feature data corresponding to the structured light image is determined using the photoluminescence difference data, and a defect area corresponding to the solar cell panel is determined based on the defect feature data.

[0009] Optionally, the fixed parameters of the solar panel and the position parameters of the light source are determined based on the shooting position, including:

[0010] Determining the shooting parameters of the camera according to the distance between the shooting position and the preset fixed position;

[0011] Using the shooting parameters to determine multiple fixed angles corresponding to the solar panel, and using the fixed angles to determine multiple irradiation angles corresponding to the light source;

[0012] Fixed parameters corresponding to the solar cell panel are determined based on the fixed angle, and position parameters corresponding to the light source are determined based on the irradiation angle.

[0013] Optionally, after controlling the camera to photograph the solar panel using fixed parameters and position parameters, a structured light image corresponding to the solar panel is obtained, including:

[0014] Acquire a first fixed angle and a second fixed angle corresponding to the solar cell panel according to the fixed parameters, and acquire a plurality of irradiation angles corresponding to the light source according to the position parameters;

[0015] After the solar panel is fixed at a fixed position by controlling the first fixed angle, the light source is controlled to illuminate the solar panel in sequence according to the illumination angle, and the camera is controlled to photograph the solar panel to obtain a first image group;

[0016] After the solar panel is fixed at a fixed position by controlling the second fixed angle, the light source is controlled to illuminate the solar panel in sequence according to the illumination angle, and the camera is controlled to photograph the solar panel to obtain a second image group;

[0017] A structured light image corresponding to the solar panel is determined based on the first image group and the second image group.

[0018] Optionally, determining the structured light image corresponding to the solar panel based on the first image group and the second image group includes:

[0019] Acquire digital images corresponding to the solar panel in the first image group and the second image group, and determine filter parameters corresponding to the digital images;

[0020] Determine a photoluminescent image region corresponding to the digital image according to the filtering parameters, and determine a dimensional parameter corresponding to the photoluminescent image region using a shooting time corresponding to the digital image;

[0021] After splicing the photoluminescence image areas according to the dimensional parameters, the structured light image corresponding to the solar panel is obtained.

[0022] Optionally, a shared feature area contained in the structured light image is extracted according to material parameters and lighting parameters of the solar panel, including:

[0023] Determine the photoluminescence characteristic parameters corresponding to the solar panel using material parameters and lighting parameters;

[0024] Acquire a digital image contained in the structured light image, and extract corresponding shared feature data in the digital image using the photoluminescence feature parameters;

[0025] A shared feature region included in the structured light image is determined based on the shared feature data.

[0026] Optionally, determining photoluminescence difference data corresponding to the structured light image using the shared feature area includes:

[0027] determining a structured light image region corresponding to the structured light image;

[0028] After eliminating the shared feature area contained in the structured light image area, the corresponding photoluminescence difference area in the structured light image area is obtained;

[0029] Photoluminescence difference data corresponding to the structured light image is determined based on the photoluminescence difference area.

[0030] Optionally, using the photoluminescence difference data to determine defect feature data corresponding to the structured light image includes:

[0031] Determine the hidden crack feature data, the broken grid feature data and the fragment feature data corresponding to the digital image in the structured light image based on the photoluminescence difference data;

[0032] Determine the deep crack data and heating area data corresponding to the digital image using the hidden crack feature data, the broken grid feature data and the fragment feature data;

[0033] Defect feature data corresponding to the structured light image is determined based on hidden crack feature data, broken grid feature data, fragment feature data, deep crack data and heating area data.

[0034] Optionally, determining a defect area corresponding to the solar panel based on the defect feature data includes:

[0035] Calculate the convolution value corresponding to the defect feature data using a preset convolution kernel, and determine the defect probability value corresponding to the defect feature data based on the convolution value;

[0036] Obtaining a binary image corresponding to the digital image in the structured light image, and determining defect position data contained in the binary image according to the defect probability value;

[0037] The defect area corresponding to the solar panel is obtained based on the defect position data.

[0038] In a second aspect, the present invention provides a solar panel defect detection system using structured light imaging, the system comprising:

[0039] An initialization unit, used to obtain a solar panel at a preset fixed position, determine a camera and a light source corresponding to the solar panel, and determine a shooting position of the camera according to the fixed position;

[0040] A shooting execution unit, used to determine fixed parameters of the solar panel and position parameters of the light source based on the shooting position, and use the fixed parameters and position parameters to control the camera to shoot the solar panel to obtain a structured light image corresponding to the solar panel;

[0041] A difference acquisition unit, used to extract a shared feature area contained in the structured light image according to material parameters and lighting parameters of the solar cell panel, and determine photoluminescence difference data corresponding to the structured light image using the shared feature area;

[0042] The defect recognition unit is used to determine defect feature data corresponding to the structured light image using the photoluminescence difference data, and determine the defect area corresponding to the solar cell panel based on the defect feature data.

[0043] In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the steps of the solar panel defect detection method using structured light imaging provided in the first aspect.

[0044] In a fourth aspect, an embodiment of the present invention further provides a storage medium storing computer executable instructions. When the computer executable instructions are called and executed by a processor, the computer executable instructions prompt the processor to implement the steps of the solar panel defect detection method using structured light imaging provided in the first aspect.

[0045] A solar panel defect detection method and system provided by the embodiment of the present invention, in the process of solar panel defect detection, first obtain a solar panel at a preset fixed position, determine the camera and light source corresponding to the solar panel, and determine the shooting position of the camera according to the fixed position; then determine the fixed parameters of the solar panel and the position parameters of the light source based on the shooting position, and use the fixed parameters and position parameters to control the camera to shoot the solar panel, and obtain the structured light image corresponding to the solar panel; then extract the shared feature area contained in the structured light image according to the material parameters and lighting parameters of the solar panel, and use the shared feature area to determine the photoluminescence difference data corresponding to the structured light image; finally, use the photoluminescence difference data to determine the defect feature data corresponding to the structured light image, and determine the defect area corresponding to the solar panel based on the defect feature data. After irradiating the solar panel with light sources from multiple angles, the method uses a camera at a fixed position to obtain the structured light image corresponding to the solar panel, and uses the structured light image to accurately obtain the photoluminescence difference data corresponding to the solar panel, thereby realizing a high-precision defect detection process in all scenarios.

[0046] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A flow chart of a method for detecting defects in a solar panel using structured light imaging provided by an embodiment of the present invention;

[0050] Figure 2 A flowchart of determining fixed parameters of a solar panel and position parameters of a light source based on a shooting position in step S102 of a method for detecting defects of a solar panel using structured light imaging provided by an embodiment of the present invention;

[0051] Figure 3In step S102 of a method for detecting defects of a solar panel using structured light imaging provided by an embodiment of the present invention, a flow chart of obtaining a structured light image corresponding to the solar panel after controlling a camera to photograph the solar panel using fixed parameters and position parameters;

[0052] Figure 4 A flowchart of step S304 of a method for detecting defects in a solar panel using structured light imaging provided by an embodiment of the present invention;

[0053] Figure 5 A flowchart of extracting a shared feature area contained in a structured light image according to material parameters and lighting parameters of a solar panel in step S103 of a method for detecting defects of a solar panel using structured light imaging provided by an embodiment of the present invention;

[0054] Figure 6 A flowchart of determining photoluminescence difference data corresponding to a structured light image using a shared feature area in step S103 of a method for detecting defects in a solar panel using structured light imaging provided by an embodiment of the present invention;

[0055] Figure 7 A flowchart of determining defect feature data corresponding to a structured light image using photoluminescence difference data in step S104 of a method for detecting defects in a solar panel using structured light imaging provided by an embodiment of the present invention;

[0056] Figure 8 A flowchart of determining a defect area corresponding to a solar panel based on defect feature data in step S104 of a method for detecting defects of a solar panel using structured light imaging provided by an embodiment of the present invention;

[0057] Fig. 9 A schematic diagram of a solar panel defect detection method using structured light imaging provided by an embodiment of the present invention;

[0058] Fig.10 A flow chart of another method for detecting defects in a solar panel using structured light imaging provided by an embodiment of the present invention;

[0059] Fig.11 A schematic diagram of the structure of a solar panel defect detection system using structured light imaging provided by an embodiment of the present invention;

[0060] Fig.12 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0061] icon:

[0062] 1110 - initialization unit; 1120 - shooting execution unit; 1130 - difference acquisition unit; 1140 - defect recognition unit;

[0063] 101 - processor; 102 - memory; 103 - bus; 104 - communication interface. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. Obviously, the described embodiments are 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.

[0065] As a clean and renewable form of energy, solar energy is being widely used in power generation systems around the world. In these systems, silicon photovoltaic panels, or solar panels, are core components that undertake the key task of converting solar energy into electrical energy. However, due to factors such as manufacturing processes, transportation, installation, and actual operating environment, the surface and interior of solar panels often have invisible defects that are difficult to detect with the naked eye, such as hidden cracks, fragments, broken grids, surface pollution, local heating, etc. These defects not only reduce the photoelectric conversion efficiency and service life of solar panels, but may also pose a safety hazard to the stable operation of the entire solar power generation system.

[0066] Existing defect detection technologies for solar panels mainly include voltammetric (IV) curve analysis, infrared thermal imaging, electroluminescence imaging, and photoluminescence imaging. These technologies have their own advantages and can be used in different scenarios. Among them, electroluminescence imaging and photoluminescence imaging technology have become a research hotspot in the current field of solar panel inspection due to their high sensitivity, high resolution, and ability to detect multiple defect types.

[0067] Electroluminescence imaging technology applies a reverse voltage to the PN junction of the photovoltaic module, causing it to emit light of a specific wavelength to present invisible defect features in the image; photoluminescence imaging uses a laser or LED light source of a specific wavelength to excite the solar panel to achieve defect imaging. The photoluminescence spectrum of silicon crystal materials is mainly concentrated in the short-wave infrared range of 1050nm to 1250nm. Due to the distortion of the lattice structure, the photoluminescence characteristics of the defective area are different from those of the normal area, which is manifested as grayscale or color changes in the imaging. These technologies have achieved good results under indoor or darkroom conditions, but because photovoltaic panels are mostly used in outdoor environments, the defect detection effect is not good under high irradiance conditions.

[0068] In addition, electroluminescence imaging also has limitations. Since electroluminescence requires voltage to be applied to the components, its detection process depends on the metallization degree and manufacturing process of the solar panel. Current injection will lead to uneven distribution of carrier concentration, affecting the luminescence intensity, thereby reducing image quality. In addition, detection requires direct contact with the solar panel components, increasing the risk of component damage or contamination.

[0069] Based on this, the present invention provides a method and system for solar panel defect detection using structured light imaging. The method uses a camera at a fixed position to obtain a structured light image corresponding to the solar panel after irradiating the solar panel with light sources from multiple angles, and uses the structured light image to accurately obtain photoluminescence difference data corresponding to the solar panel, thereby realizing a high-precision defect detection process in all scenarios.

[0070] To facilitate understanding of this embodiment, a method for detecting defects in a solar panel using structured light imaging disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, including:

[0071] Step S101, obtaining a solar panel at a preset fixed position, determining a camera and a light source corresponding to the solar panel, and determining a shooting position of the camera according to the fixed position.

[0072] In the specific implementation process, a fixed position can be pre-set on the production line of solar panels or on related testing equipment. The selection of this position needs to comprehensively consider factors such as operational convenience and detection accuracy. Then the solar panel to be tested is accurately placed on this preset fixed position. This step is the basis for subsequent testing work. The accuracy of this position will affect the quality of the captured image and the accuracy of subsequent analysis.

[0073] Next, we need to determine the camera and light source that are compatible with the solar panel. The choice of camera should be based on factors such as the size of the solar panel and the detection accuracy requirements. For example, for larger panels that require high-precision detection, industrial cameras with high resolution and large target areas should be used to obtain high-sensitivity infrared imaging, so that the subtle features on the surface of the panel can be clearly captured. The choice of light source is also very critical. Different types of light sources will produce different photoluminescence effects. According to the material and surface characteristics of the solar panel, the light source with appropriate intensity and irradiation method should be selected to ensure that the key features are highlighted when the image is captured.

[0074] Finally, according to the fixed position of the solar panel, the camera's shooting position is reasonably determined. This process needs to consider the distance and angle between the camera and the panel. Generally speaking, the camera should be shot perpendicular to the surface of the panel as much as possible to reduce image distortion, and ensure that the shooting range can fully cover the panel, while avoiding blurry or unclear images due to being too close or too far.

[0075] Step S102, determining fixed parameters of the solar panel and position parameters of the light source based on the shooting position, and controlling the camera to shoot the solar panel using the fixed parameters and the position parameters to obtain a structured light image corresponding to the solar panel.

[0076] Based on the camera shooting position determined above, the fixed parameters of the solar panel and the position parameters of the light source are further clarified. The fixed parameters of the solar panel include its specific coordinates in the horizontal and vertical directions, tilt angle, etc. The position parameters of the light source are mainly used to characterize the illumination direction between the light source and the solar panel, etc., while the fixed parameters are mainly used to characterize the placement of the solar panel in a fixed position.

[0077] After the solar panel is fixed using these fixed parameters, the position parameters are used to control the light source to illuminate the solar panel, and then the camera is controlled to shoot the solar panel. Since the position parameters involve multiple lighting directions and correspond to multiple different illumination angles, the digital images corresponding to the solar panel taken at all illumination angles are integrated to obtain the corresponding structured light image.

[0078] Step S103 : extracting a shared feature region contained in the structured light image according to the material parameters and lighting parameters of the solar cell panel, and determining photoluminescence difference data corresponding to the structured light image using the shared feature region.

[0079] According to the material parameters of the solar panel (such as silicon purity, crystal structure, etc.) and lighting parameters (such as light intensity, uniformity, etc.), the captured structured light image is deeply analyzed to extract the shared feature areas contained therein. The shared feature areas refer to the areas with similar optical properties and texture features corresponding to the surface of the solar panel in the structured light image. The characteristics of these areas can reflect the normal structure and performance of the solar panel.

[0080] Using these shared feature regions, the photoluminescence difference data corresponding to the structured light image is further determined. Photoluminescence refers to the light reflected by the solar panel under the illumination of a light source. There will be differences in the intensity and distribution of photoluminescence between normal areas and defective areas. By comparing the photoluminescence conditions at different locations within the shared feature region, the photoluminescence difference data is calculated, which can serve as an important basis for determining whether the solar panel has defects.

[0081] Step S104 : Determine defect feature data corresponding to the structured light image using the photoluminescence difference data, and determine a defect region corresponding to the solar cell panel based on the defect feature data.

[0082] The photoluminescence difference data obtained above is used for in-depth analysis and processing to determine the defect feature data corresponding to the structured light image. The specific implementation process can be achieved through a machine learning algorithm or a preset threshold judgment method. For example, the photoluminescence difference data is classified and identified using a trained convolutional neural network model to determine the defect type (such as cracks, holes, impurities, etc.), size, shape and other feature information.

[0083] Based on the determined defect feature data, the defect area corresponding to the solar panel is accurately located in the structured light image. In actual scenes, the defect area can be separated from the normal area through the image segmentation algorithm, and the specific location and range of the defect can be marked.

[0084] Optionally, the fixed parameters of the solar panel and the position parameters of the light source are determined based on the shooting position, such as Figure 2 As shown, including:

[0085] Step S201, determining the shooting parameters of the camera according to the distance between the shooting position and the preset fixed position.

[0086] In actual operation, the distance between the shooting position and the preset fixed position is a key factor that significantly affects the camera's ability to obtain clear and accurate photofluorescence images. When the distance between the shooting position and the preset fixed position changes, the camera needs to adjust its shooting parameters accordingly to ensure image quality.

[0087] Step S202: using the shooting parameters to determine a plurality of fixed angles corresponding to the solar cell panel, and using the fixed angles to determine a plurality of irradiation angles corresponding to the light source.

[0088] After the shooting parameters are determined, multiple fixed angles corresponding to the solar panel can be determined based on this to fully capture the various features of the panel. By adjusting the fixed angle of the solar panel, the camera can shoot the solar panel from different angles to obtain more comprehensive information. After determining the fixed angle of the solar panel, the multiple illumination angles corresponding to the light source can be further determined. The illumination angle of the light source is crucial to highlighting the features of the panel. Under different fixed angles, different illumination angles of the light source need to be set to obtain multiple illumination angles corresponding to the light source.

[0089] Step S203, determining fixed parameters corresponding to the solar cell panel based on the fixed angle, and determining position parameters corresponding to the light source based on the irradiation angle.

[0090] According to the fixed angle of the solar panel determined in the above steps, its corresponding fixed parameters can be further determined. These fixed parameters include the specific position coordinates of the panel on the detection platform and the precise value of the rotation angle. By precisely controlling these fixed parameters, the position and posture of the panel are highly consistent during each detection, thereby improving the accuracy and repeatability of the detection results. Similarly, based on the irradiation angle of the light source, the position parameters corresponding to the light source can be determined, including the distance between the light source and the panel, the horizontal and vertical offsets, etc. By precisely adjusting the position parameters of the light source, precise control of the irradiation angle of the panel can be achieved.

[0091] Optionally, after the camera is controlled to shoot the solar panel using fixed parameters and position parameters, a structured light image corresponding to the solar panel is obtained, such as Figure 3 As shown, including:

[0092] Step S301, obtaining a first fixed angle and a second fixed angle corresponding to the solar cell panel according to fixed parameters, and obtaining a plurality of irradiation angles corresponding to the light source according to position parameters.

[0093] After determining the fixed parameters of the solar panel and the position parameters of the light source, the first fixed angle and the second fixed angle corresponding to the solar panel can be extracted from the fixed parameters. These two fixed angles are pre-set according to the detection requirements and the characteristics of the solar panel. Different angles allow the camera to capture information about the panel from different perspectives, which helps to detect defects in the panel more comprehensively. For example, the solar panel at the first fixed angle is in a horizontal state, and the solar panel at the second fixed angle is obtained by rotating the solar panel at the first fixed angle 90 degrees along the center point, and it is still in a horizontal state.

[0094] At the same time, multiple illumination angles corresponding to the light source are set according to the position parameters of the light source, such as 0°, 45°, 90°, 135° and 180° (the angle is the angle between the illumination direction of the light source and the vertical line of the solar panel). These illumination angles are set to illuminate the solar panel from different directions, so that the features of the panel surface can be clearly displayed under different lighting conditions.

[0095] Step S302, after the solar panel is fixed at a fixed position by controlling the first fixed angle, the light source is controlled to illuminate the solar panel in sequence according to the illumination angle, and the camera is controlled to photograph the solar panel to obtain a first image group.

[0096] After the solar panel is fixed at a fixed position by controlling the first fixed angle, the solar panel is illuminated at the five illumination angles respectively, and then the camera is controlled to shoot the solar panel to obtain five images, which are set as the first image group.

[0097] Step S303, after the solar panel is fixed at a fixed position by controlling the second fixed angle, the light source is controlled to illuminate the solar panel in sequence according to the illumination angle, and the camera is controlled to photograph the solar panel to obtain a second image group.

[0098] After the first image group is acquired, the solar panel is fixed at a fixed position by controlling the second fixed angle, and then the solar panel is illuminated according to the above five illumination angles respectively, and then the camera is controlled to shoot the solar panel to obtain another five images, which are set as the second image group.

[0099] Step S304: determining a structured light image corresponding to the solar panel based on the first image group and the second image group.

[0100] The first image group and the second image group have a total of 10 images, which can be combined and processed accordingly to obtain the structured light image corresponding to the solar panel. Specifically, the step S304 of determining the structured light image corresponding to the solar panel based on the first image group and the second image group is as follows: Figure 4 As shown, including:

[0101] Step S401, acquiring digital images corresponding to the solar cell panels in the first image group and the second image group, and determining filter parameters corresponding to the digital images;

[0102] Step S402, determining a photoluminescent image region corresponding to the digital image according to the filtering parameters, and determining a dimensional parameter corresponding to the photoluminescent image region using a shooting time corresponding to the digital image;

[0103] Step S403 , after splicing the photoluminescence image regions according to the dimensional parameters, a structured light image corresponding to the solar cell panel is obtained.

[0104] The generation process of the structured light image relies on the digital images in the first image group and the second image group. After acquiring the corresponding digital images, the filter parameters corresponding to the solar panel are determined to obtain the photoluminescent area corresponding to the digital image.

[0105] Another key parameter in the structured light image is the dimension parameter, which represents the specific position of the digital image in the structured light image. The generation process of the dimension parameter can be realized by using the shooting time corresponding to the digital image, that is, the earlier the digital image is shot, the smaller the value of the corresponding dimension parameter is, and the later the digital image is shot, the larger the value of the corresponding dimension parameter is.

[0106] In layman's terms, the dimension parameter is a unique marking value of the photoluminescent image area in the digital image. Therefore, the dimension parameter can be used to stitch the photoluminescent image areas corresponding to all digital images to obtain the structured light image corresponding to the solar panel.

[0107] Optionally, the shared feature area contained in the structured light image is extracted according to the material parameters and lighting parameters of the solar panel, such as Figure 5 As shown, including:

[0108] Step S501, determining the photoluminescence characteristic parameters corresponding to the solar cell panel using the material parameters and the lighting parameters.

[0109] The material parameters and lighting parameters of solar panels are key factors in determining the photoluminescence characteristic parameters. In terms of material parameters, different types of solar panel materials have significantly different photoluminescence characteristics. Lighting parameters cannot be ignored either. For example, light intensity directly affects the intensity of photoluminescence. Generally speaking, the higher the light intensity, the greater the stimulated photoluminescence intensity. The uniformity of light is also very important. If the lighting is uneven, the photoluminescence intensity of different areas of the panel will be inconsistent, affecting the subsequent feature extraction. The spectral characteristics of light will also interact with the material of the panel. Light of different spectra may stimulate different photoluminescence responses of the panel. By comprehensively analyzing and experimentally measuring the material parameters and lighting parameters, the corresponding photoluminescence characteristic parameters of the solar panel can be determined.

[0110] Step S502 : acquiring a digital image contained in the structured light image, and extracting corresponding shared feature data in the digital image using the photoluminescence feature parameters.

[0111] The digital image is analyzed and processed using the photoluminescence characteristic parameters determined in the above steps to extract the corresponding shared characteristic data, that is, the area without defects. Specifically, the photoluminescence characteristic parameters provide an important basis for feature extraction. For example, according to the peak wavelength of photoluminescence, the corresponding pixel points can be screened out in the spectrum information of the digital image. These pixel points may represent the normal luminous area of ​​the solar panel, that is, the area without defects.

[0112] Step S503: determining a shared feature region contained in the structured light image based on the shared feature data.

[0113] The shared feature data can characterize the area with similar optical properties and texture features corresponding to the surface of the solar panel in the structured light image. Therefore, the shared feature area can be determined by obtaining the position parameters corresponding to the area. The solar panel in the shared feature area has normal structure and performance and does not contain related defects.

[0114] Optionally, the photoluminescence difference data corresponding to the structured light image is determined using the shared feature area, such as Figure 6 As shown, including:

[0115] Step S601: determine a structured light image area corresponding to the structured light image.

[0116] A structured light image is an image obtained by photographing a solar panel, which contains the photoluminescence information of each part of the panel. To determine the structured light image area corresponding to the structured light image, it is first necessary to fully define the entire structured light image, clarify the boundary range of the image, and cover all the photographed panel areas.

[0117] In actual operation, there may be some edge parts of the image that may have inaccurate or incomplete information due to factors such as shooting angle and lighting conditions. Therefore, when determining the structured light image area, it is necessary to pre-process the image, such as cropping off those obviously abnormal or invalid edge parts, to ensure that the determined structured light image area can accurately reflect the actual situation of the solar panel. At the same time, the image segmentation algorithm can also be used to distinguish the panel area from the background area in the structured light image based on the gray value, color and other features of the image, so as to more accurately determine the structured light image area.

[0118] Step S602 : After eliminating the shared feature region contained in the structured light image region, the corresponding photoluminescence difference region in the structured light image region is obtained.

[0119] The shared feature areas represent the photoluminescence characteristics of the solar panel under normal conditions. In order to find out the areas where defects may exist, these shared feature areas need to be removed from the structured light image area. When removing the shared feature areas, the corresponding marking and removal operations can be performed in the structured light image area based on the position and range information obtained when the shared feature areas were previously extracted. The remaining areas are the areas where photoluminescence differences may exist, that is, the photoluminescence difference areas.

[0120] The photoluminescence difference area may be due to defects in the solar panel (such as cracks, impurities, etc.), which causes the photoluminescence characteristics of the area to be different from the normal shared feature area. These difference areas may appear as abnormalities in brightness, color, texture, etc. in the image, and further analysis is needed to determine the specific photoluminescence difference.

[0121] Step S603 : determining photoluminescence difference data corresponding to the structured light image based on the photoluminescence difference area.

[0122] After the photoluminescence difference areas are determined, the photoluminescence difference data corresponding to the structured light image can be calculated based on these areas. For example, the photoluminescence intensity, color and other characteristics of each pixel in the photoluminescence difference area can be measured and counted, and the average photoluminescence intensity, maximum photoluminescence intensity, minimum photoluminescence intensity and other parameters in the area, as well as the color distribution, can be calculated. The photoluminescence difference data can intuitively reflect the differences in the photoluminescence characteristics of solar panels in different areas, providing an important basis for subsequent judgment of whether there are defects in the solar panels and the type and degree of the defects.

[0123] Optionally, the defect feature data corresponding to the structured light image is determined using the photoluminescence difference data, such as Figure 7 As shown, including:

[0124] Step S701 : determining the hidden crack feature data, the broken grid feature data and the fragment feature data corresponding to the digital image in the structured light image based on the photoluminescence difference data.

[0125] Photoluminescence difference data reflects the differences in photoluminescence characteristics of different areas of solar panels, and these differences are often closely related to defects in the panels. By carefully analyzing the photoluminescence difference data, characteristic data corresponding to defects such as hidden crack features, broken grid features and fragment features can be extracted from them.

[0126] Hidden cracks are common and difficult to detect defects in solar panels. In the photoluminescence difference data, the hidden crack area usually shows a local decrease in photoluminescence intensity and presents a thin linear distribution. By performing edge detection and line extraction algorithm processing on the photoluminescence difference data, these thin and long low-luminescence intensity areas can be identified, and then the location, length and direction of the hidden cracks and other hidden crack feature data can be determined.

[0127] The grid lines of the solar panel are responsible for collecting and transmitting current. A broken grid will result in obstruction of current transmission. The broken grid area will show obvious discontinuity characteristics in the photoluminescence difference data, that is, the photoluminescence intensity at the location of the grid line will suddenly interrupt or change abnormally. Through morphological analysis and connected area detection of the photoluminescence difference data, the broken grid feature data such as the location of the broken grid, the number of breakpoints, and the length of the broken grid can be determined.

[0128] The fragmentation of the solar panel is a serious defect. The photoluminescence characteristics of the fragmented area are very different from those of the normal area. In the photoluminescence difference data, the fragmented area usually shows an abnormal decrease in photoluminescence intensity over a large area, and the boundary is relatively clear. The image segmentation algorithm can be used to segment the fragmented area from the image according to the threshold of the photoluminescence intensity, so as to determine the fragment feature data such as the area, shape and position of the fragment.

[0129] Step S702, using the hidden crack feature data, the broken grid feature data and the fragment feature data, determine the deep crack data and the heating area data corresponding to the digital image.

[0130] Hidden crack feature data, broken grid feature data and fragment feature data provide important clues for inferring deep cracks. Although deep cracks may not be easy to observe directly on the surface, they will affect the structure and performance inside the solar panel, which will be reflected in the photoluminescence difference data. For example, when the hidden crack features show a trend of multiple intertwined and deep inside the solar panel, it may indicate the existence of deep cracks. By spatially analyzing the distribution of hidden cracks, broken grids and fragments, deep crack data such as the location, depth and extension direction of deep cracks can be inferred.

[0131] Defects such as broken grids and debris will increase the local resistance of the panel and generate additional heat during the operation of the solar panel. Based on the characteristic data of hidden cracks, broken grids and debris, the area where heating problems may occur can be located. Combining the principle of thermal imaging and the relationship between photoluminescence and thermal effects, the temperature distribution, area size and location of the heating area can be estimated by further processing and analyzing the photoluminescence difference data, such as temperature inversion algorithm.

[0132] Step S703, determining defect feature data corresponding to the structured light image according to the hidden crack feature data, the broken grid feature data, the fragment feature data, the deep crack data and the heating area data.

[0133] By integrating and comprehensively analyzing the hidden crack feature data, broken grid feature data, fragment feature data, deep crack data, and heating area data obtained in the previous steps, the complete defect feature data corresponding to the structured light image can be determined. These defect feature data cover detailed information on different types of defects in the solar panel, including the location, size, shape, severity, etc. of the defects.

[0134] Optionally, the defect area corresponding to the solar panel is determined based on the defect feature data, such as Figure 8 As shown, including:

[0135] Step S801, using a preset convolution kernel to calculate a convolution value corresponding to the defect feature data, and determining a defect probability value corresponding to the defect feature data based on the convolution value.

[0136] The convolution operation can effectively extract feature information from the image. In the process of determining the defect area of ​​the solar panel, the preset convolution kernel can be used to perform convolution operation on the defect feature data. The convolution kernel can be designed according to the common defect features of solar panels. When the convolution kernel is used to perform convolution operation on the defect feature data, the convolution kernel slides on the defect feature data and performs weighted summation on the local area of ​​each position to obtain the corresponding convolution value. The convolution value reflects the degree of match between the defect feature data at different positions and the defect features represented by the convolution kernel. The larger the convolution value, the more similar the features at the position are to the preset defect features. Based on these convolution values, they can be converted into defect probability values ​​through certain mapping functions.

[0137] Step S802 , obtaining a binary image corresponding to the digital image in the structured light image, and determining defect position data contained in the binary image according to the defect probability value.

[0138] In order to more conveniently identify and locate defective areas, the digital image in the structured light image can be converted into a binary image. A binary image has only two grayscale values, usually 0 and 255. By setting a suitable threshold, the pixels in the digital image with grayscale values ​​greater than the threshold are set to 255, and the pixels with grayscale values ​​less than the threshold are set to 0, thereby obtaining a binary image.

[0139] After obtaining the binary image, the defect location data contained in the binary image is determined in combination with the defect probability value calculated previously. For each pixel in the binary image, a judgment can be made based on its corresponding defect probability value. If the defect probability value exceeds a preset threshold, it is considered that there may be a defect at the location of the pixel, and it is marked as a defect location. By traversing all the pixels in the binary image, a series of defect location data can be obtained, which records the coordinate information of the pixels that may have defects.

[0140] Step S803: acquiring a defect area corresponding to the solar cell panel based on the defect position data.

[0141] After obtaining the defect location data, the defect area corresponding to the solar panel can be further determined. In the specific implementation process, the defect location data can be clustered and the adjacent and close defect location points can be classified into the same category. Each cluster represents a possible defect area. Then, the specific range of the defect area is determined by calculating the boundaries of all defect location points in each cluster. The minimum circumscribed rectangle, convex hull and other methods can be used to calculate the boundaries of the defect area. Finally, according to the determined boundaries, the corresponding defect areas are marked on the structured light image or the actual solar panel. These defect areas intuitively show the possible defective parts of the solar panel, providing a clear target for subsequent repairs or quality assessments.

[0142] like Fig. 9 The schematic diagram of a method for defect detection of solar panels using structured light imaging is shown. First, the photovoltaic panel is illuminated by an excitation light source arranged at multiple angles, and a photoluminescence image is obtained in combination with a high-sensitivity imaging device. The multi-angle illumination design can change the excitation path of photoluminescence, highlight the differences in photoluminescence characteristics caused by defects (such as hidden cracks, broken grids, fragments, etc.), thereby significantly enhancing the visualization of defects and avoiding detection blind spots caused by single-angle illumination. After completing multi-angle illumination and imaging, the position of the light source is adjusted for a second round of illumination and imaging to further stimulate potential complex defects of the photovoltaic panel from another spatial direction. The multi-position arrangement of excitation light sources can avoid the omission of defect information due to the single incident path of the light source in a specific area, and comprehensively capture defect characteristics through multi-dimensional perspectives to achieve accurate detection of complex defects (such as deep cracks, local heating areas, etc.).

[0143] like Fig.10 The schematic diagram of the solar panel defect detection method using structured light imaging with convolutional neural network is shown in FIG. In order to more efficiently identify defects in structured light images, Fig.10 The convolutional neural network used in the paper is a structured light image defect recognition network, which contains 9 defect enhancement units. The design of this unit is based on the characteristics of structured light imaging, that is, during the shooting process, the positions of the camera and solar panel remain unchanged, and only the position and angle of the light source are changed. This arrangement makes the sunlight an invariant in a short period of time, which provides key inspiration for the development of the defect enhancement unit.

[0144] The defect enhancement unit first extracts its shared features (mainly dominated by factors such as surface material and uniform illumination) from the structured light image group. Subsequently, by performing a subtraction operation on the shared features, the photoluminescence differences caused by changes in the position and angle of the light source are separated, and the defect areas exposed by changes in the excitation conditions are accurately located. This method solves the problem that a single light source cannot excite certain defects, and realizes the effective fusion of defect information under the excitation of different light sources. Based on the autoencoder architecture, the defect enhancement unit extracts and enhances defect features through the encoding and decoding process. The processed features are input into the prediction head composed of convolutional layers, Softmax and binarization operations, and finally high-precision detection and classification of defects are achieved. Combined with the photoluminescence images obtained under multi-angle and multi-position excitation conditions, the unit can further improve the comprehensiveness and accuracy of defect detection. The deep learning network uses the multi-dimensional characteristics of photoluminescence images to systematically analyze and classify the type, location and range of defects. This method fully reveals the complex defect characteristics of photovoltaic panels and significantly improves the detection efficiency.

[0145] Combination Fig. 9 and Fig.10 In the specific scenario described in, the solar panel defect detection method includes the following steps:

[0146] Step 1: Adjust the light source to positions with angles of 0°, 45°, 90° (perpendicular to the solar panel), 135°, and 180° to the solar panel, illuminate the solar panel in turn, take a photoluminescent image at each angle, and obtain a total of 5 images.

[0147] Step 2: After completing the irradiation and shooting at the above fixed angle, adjust the position of the light source to rotate it 90° around the solar panel (such as moving from the original side of the solar panel to the adjacent side), and then irradiate and shoot again from 0° to 180°. Repeat this cycle to obtain a total of 10 images. During each irradiation, a high-sensitivity near-infrared camera is used for imaging, and a wavelength-selective filter is used to ensure that the imaging system only captures the photoluminescence signal.

[0148] Step 3: Send the 10 photos taken into the designed structured light imaging defect recognition network for defect recognition.

[0149] Step 4: Dimensional stitching of the 10 acquired photoluminescence images:

[0150] Step 5: Input the spliced ​​image into the defect enhancement unit and extract the shared features of the image group through convolution operation:

[0151] Step 6: Subtract the shared features from the original structured light image to highlight the defect area, extract the spatial defect features through spatial convolution, and further optimize the feature fusion by combining point convolution to obtain the output of the first defect enhancement unit:

[0152] Step 7: Perform encoding and decoding of the network according to the design flow chart, extract features through 9 defect enhancement units layer by layer, and finally obtain the defect enhancement result in the ninth defect enhancement unit. Use the defect enhancement result to calculate the specific defect location through the prediction head of convolution, Softmax and binarization operations.

[0153] The model training can be supervised by using the cross entropy loss function, and a holographic defect data set containing different defect types and distributions is constructed to optimize the training weights. After the training is completed, the optimized weights are loaded into the detection network, and the defect detection task is performed according to the process from the first step to the seventh step to realize the defect recognition of solar panels based on structured light imaging.

[0154] From the solar panel defect detection method using structured light imaging mentioned in the above embodiments, it can be seen that after illuminating the solar panel with light sources from multiple angles, the method uses a camera at a fixed position to obtain a structured light image corresponding to the solar panel, and uses the structured light image to accurately obtain the photoluminescence difference data corresponding to the solar panel, thereby realizing a high-precision defect detection process in all scenarios.

[0155] Corresponding to the solar panel defect detection method using structured light imaging provided in the above-mentioned embodiment, an embodiment of the present invention provides a solar panel defect detection system using structured light imaging, such as Fig.11 As shown, the system includes:

[0156] The initialization unit 1110 is used to obtain a solar panel at a preset fixed position, determine a camera and a light source corresponding to the solar panel, and determine a shooting position of the camera according to the fixed position;

[0157] The shooting execution unit 1120 is used to determine the fixed parameters of the solar panel and the position parameters of the light source based on the shooting position, and use the fixed parameters and the position parameters to control the camera to shoot the solar panel to obtain a structured light image corresponding to the solar panel;

[0158] The difference acquisition unit 1130 is used to extract the shared feature area contained in the structured light image according to the material parameters and lighting parameters of the solar cell panel, and determine the photoluminescence difference data corresponding to the structured light image using the shared feature area;

[0159] The defect recognition unit 1140 is used to determine defect feature data corresponding to the structured light image using the photoluminescence difference data, and determine a defect area corresponding to the solar cell panel based on the defect feature data.

[0160] From the structured light imaging solar panel defect detection system mentioned in the above embodiments, it can be seen that after the system illuminates the solar panel with light sources from multiple angles, it uses a camera at a fixed position to obtain the structured light image corresponding to the solar panel, and uses the structured light image to accurately obtain the photoluminescence difference data corresponding to the solar panel, thereby realizing a high-precision defect detection process in all scenarios.

[0161] The solar panel defect detection system using structured light imaging provided in the embodiment of the present invention has the same implementation principle and technical effects as those in the aforementioned solar panel defect detection method using structured light imaging embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference may be made to the corresponding contents in the aforementioned solar panel defect detection method using structured light imaging embodiment.

[0162] This embodiment also provides an electronic device. The structural diagram of the electronic device is as follows: Fig.12As shown, the device includes a processor 101 and a memory 102; wherein the memory 102 is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the steps of the above-mentioned solar panel defect detection method using structured light imaging.

[0163] Fig.12 The electronic device shown further includes a bus 103 and a communication interface 104 , and the processor 101 , the communication interface 104 and the memory 102 are connected via the bus 103 .

[0164] The memory 102 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The bus 103 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.12 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0165] The communication interface 104 is used to connect to at least one user terminal and other network units through a network interface, and send the encapsulated IPv4 message or IPv4 message to the user terminal through the network interface.

[0166] The processor 101 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 101. The above processor 101 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 102, and the processor 101 reads the information in the memory 102 and completes the steps of the method of the above embodiment in combination with its hardware.

[0167] An embodiment of the present invention further provides a storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for detecting defects of a solar panel using structured light imaging in the aforementioned embodiment are executed.

[0168] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, equipment and methods can be implemented in other ways. The system embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0169] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0171] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present invention can essentially or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0172] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for detecting defects in solar panels using structured light imaging, characterized in that: The method comprises: Acquire a solar panel at a preset fixed position, determine a camera and a light source corresponding to the solar panel, and determine a shooting position of the camera according to the fixed position; Determining fixed parameters of the solar panel and position parameters of the light source based on the shooting position, and controlling the camera to shoot the solar panel using the fixed parameters and the position parameters to obtain a structured light image corresponding to the solar panel; Extracting a shared feature area contained in the structured light image according to material parameters and lighting parameters of the solar cell panel, and determining photoluminescence difference data corresponding to the structured light image using the shared feature area; Defect feature data corresponding to the structured light image is determined using the photoluminescence difference data, and a defect area corresponding to the solar cell panel is determined based on the defect feature data.

2. The method for detecting solar panel defects by structured light imaging according to claim 1, characterized in that: Determining fixed parameters of the solar panel and position parameters of the light source based on the shooting position includes: Determining a shooting parameter of the camera according to a distance between the shooting position and the preset fixed position; Determine a plurality of fixed angles corresponding to the solar cell panel using the shooting parameters, and determine a plurality of irradiation angles corresponding to the light source using the fixed angles; The fixed parameter corresponding to the solar cell panel is determined based on the fixed angle, and the position parameter corresponding to the light source is determined based on the irradiation angle.

3. The method for detecting solar panel defects by structured light imaging according to claim 2, characterized in that: After the camera is controlled to photograph the solar panel using the fixed parameters and the position parameters, a structured light image corresponding to the solar panel is obtained, including: Acquire a first fixed angle and a second fixed angle corresponding to the solar cell panel according to the fixed parameters, and acquire a plurality of irradiation angles corresponding to the light source according to the position parameters; After the solar cell panel is fixed at the fixed position by using the first fixed angle, the light source is controlled in sequence to illuminate the solar cell panel according to the illumination angle, and the camera is controlled to photograph the solar cell panel to obtain a first image group; After the solar panel is fixed at the fixed position by controlling the second fixed angle, the light source is sequentially controlled to illuminate the solar panel according to the illumination angle, and the camera is controlled to photograph the solar panel to obtain a second image group; The structured light image corresponding to the solar panel is determined based on the first image group and the second image group.

4. The method for detecting solar panel defects by structured light imaging according to claim 3, characterized in that: Determining the structured light image corresponding to the solar panel based on the first image group and the second image group includes: Acquire digital images corresponding to the solar panel in the first image group and the second image group, and determine filter parameters corresponding to the digital images; Determine a photoluminescent image region corresponding to the digital image according to the filtering parameters, and determine a dimensional parameter corresponding to the photoluminescent image region using a shooting time corresponding to the digital image; After the photoluminescence image regions are spliced ​​according to the dimensional parameters, the structured light image corresponding to the solar cell panel is obtained.

5. The solar panel defect detection method based on structured light imaging according to claim 1, characterized in that: Extracting a shared feature area contained in the structured light image according to the material parameters and lighting parameters of the solar panel includes: Determining the photoluminescence characteristic parameters corresponding to the solar cell panel using the material parameters and the lighting parameters; Acquire a digital image contained in the structured light image, and extract corresponding shared feature data in the digital image using the photoluminescence feature parameters; The shared feature area included in the structured light image is determined based on the shared feature data.

6. The method for detecting solar panel defects using structured light imaging according to claim 5, characterized in that: Determining photoluminescence difference data corresponding to the structured light image using the shared feature area includes: determining a structured light image area corresponding to the structured light image; After eliminating the shared feature area contained in the structured light image area, obtaining a corresponding photoluminescence difference area in the structured light image area; The photoluminescence difference data corresponding to the structured light image is determined based on the photoluminescence difference area.

7. The solar panel defect detection method based on structured light imaging according to claim 1, characterized in that: Determining defect feature data corresponding to the structured light image using the photoluminescence difference data includes: Determine, based on the photoluminescence difference data, hidden crack feature data, broken grid feature data, and fragment feature data corresponding to the digital image in the structured light image; Determine the deep crack data and heating area data corresponding to the digital image by using the hidden crack feature data, the broken grid feature data and the fragment feature data; The defect feature data corresponding to the structured light image is determined according to the hidden crack feature data, the broken grid feature data, the fragment feature data, the deep crack data and the heating area data.

8. The method for detecting solar panel defects using structured light imaging according to claim 1, characterized in that: Determining a defect area corresponding to the solar panel based on the defect feature data includes: Calculating a convolution value corresponding to the defect feature data using a preset convolution kernel, and determining a defect probability value corresponding to the defect feature data based on the convolution value; Acquire a binary image corresponding to the digital image in the structured light image, and determine defect position data contained in the binary image according to the defect probability value; The defect area corresponding to the solar cell panel is acquired based on the defect position data.

9. A solar panel defect detection system based on structured light imaging, characterized in that: The system comprises: an initialization unit, configured to obtain a solar panel at a preset fixed position, determine a camera and a light source corresponding to the solar panel, and determine a shooting position of the camera according to the fixed position; a shooting execution unit, configured to determine fixed parameters of the solar panel and position parameters of the light source based on the shooting position, and to control the camera to shoot the solar panel using the fixed parameters and the position parameters to obtain a structured light image corresponding to the solar panel; a difference acquisition unit, configured to extract a shared feature region contained in the structured light image according to material parameters and lighting parameters of the solar cell panel, and determine photoluminescence difference data corresponding to the structured light image using the shared feature region; The defect recognition unit is used to determine defect feature data corresponding to the structured light image using the photoluminescence difference data, and determine a defect area corresponding to the solar cell panel based on the defect feature data.

10. An electronic device, characterized in that: The invention comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the steps of the solar panel defect detection method using structured light imaging as described in any one of claims 1 to 8.

Images