Photovoltaic cell defect detection method and defect detection system

Abstract

The application relates to a photovoltaic cell defect detection method and a defect detection system. A host process acquires an image path of a newly added cell image stored on a detection device corresponding to each photovoltaic cell production line, and sends the image path to a working process. The working process acquires the newly added cell image corresponding to the image path, segments the newly added cell image based on a preset quadrant number to obtain each quadrant image corresponding to the newly added cell image, and detects bright and dark defects of the cell image according to at least one quadrant image to obtain a defect detection result of the photovoltaic cell. In the embodiment of the application, the detection server segments the newly added cell image to obtain each quadrant image corresponding to the newly added cell image, and detects bright and dark defects of the cell image according to at least one quadrant image to obtain the defect detection result of the photovoltaic cell, thereby improving the defect detection efficiency of the photovoltaic cell.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor technology, and in particular to a photovoltaic cell defect detection method and a defect detection system. BACKGROUND

[0002] The bright and dark defect detection of the cell is a crucial quality control link in the photovoltaic intelligent manufacturing, and is directly related to the product performance. Any bright and dark defect such as crack, broken grid and dirt will seriously damage the internal structure of the cell, resulting in a decrease in conversion efficiency and a loss of output power. Moreover, it will also affect the reliability and safety of the final assembly. Therefore, the bright and dark defect detection of the cell is a key cornerstone for improving product competitiveness, ensuring the safe and stable operation of the power station, and reducing the life cycle cost.

[0003] At present, a neural network is usually used to detect multiple cell images, but this method has the problems of slow detection speed and low efficiency. SUMMARY

[0004] Therefore, it is necessary to provide a photovoltaic cell defect detection method and a defect detection system capable of improving detection efficiency in view of the above technical problems.

[0005] In a first aspect, the present application provides a photovoltaic cell defect detection method, which comprises the following steps:

[0006] The main process obtains the image path of the newly added cell image stored on the detection equipment corresponding to each photovoltaic cell production line, and sends the image path to the working process.

[0007] The working process obtains the newly added cell image corresponding to the image path, segments the newly added cell image based on a preset quadrant number, and obtains each quadrant image corresponding to the newly added cell image.

[0008] The bright and dark defect detection of the cell image is performed according to at least one quadrant image to obtain the defect detection result of the photovoltaic cell.

[0009] In one embodiment, the bright and dark defect detection of the cell image is performed according to at least one quadrant image to obtain the defect detection result of the photovoltaic cell, which comprises the following steps:

[0010] The bright and dark defect detection of the cell image is performed according to the comparison result of the gray value of at least one quadrant image and a preset gray threshold to obtain the defect detection result of the photovoltaic cell.

[0011] In one embodiment, the bright and dark defect detection of the cell image is performed according to the comparison result of the gray value of at least one quadrant image and a preset gray threshold to obtain the defect detection result of the photovoltaic cell, which comprises the following steps:

[0012] Determine a first comparison result between the gray value of the first quadrant image in at least one quadrant image and a preset gray value threshold;

[0013] If the first comparison result is that the gray value of the first quadrant image is greater than a preset gray value threshold, a second comparison result is determined between the gray value of the second quadrant image in at least one quadrant image and the preset gray value threshold.

[0014] If the grayscale value of the second quadrant image is not greater than the preset grayscale threshold in the second comparison result, the second quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell. The defect detection result includes the anomaly type of the photovoltaic cell corresponding to the newly added cell image, and the comparison result includes the first comparison result and the second comparison result.

[0015] In one embodiment, the photovoltaic cell defect detection method includes:

[0016] If the grayscale value of the first quadrant image is not greater than the preset grayscale threshold in the first comparison result, the first quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell.

[0017] In one embodiment, the photovoltaic cell defect detection method further includes:

[0018] Convert each quadrant image in at least one quadrant image into an array, and then convert the array into grayscale values ​​of the quadrant images.

[0019] In one embodiment, at least one quadrant image includes quadrant images corresponding to the newly added solar cell image. Based on the comparison result of the grayscale value of the at least one quadrant image with a preset grayscale threshold, brightness and darkness defect detection is performed on the solar cell image to obtain the defect detection result of the photovoltaic solar cell, including:

[0020] If the grayscale values ​​of all quadrant images corresponding to the newly added solar cell image are greater than the preset grayscale threshold, the defect detection result of the photovoltaic solar cell is determined to be that the photovoltaic solar cell corresponding to the newly added solar cell image is normal.

[0021] In one embodiment, the photovoltaic cell defect detection method further includes:

[0022] If there is a gray value in each quadrant image corresponding to the newly added solar cell image that is not greater than a preset gray value threshold, the defect detection result of the photovoltaic solar cell is determined based on the quadrant image that is not greater than the preset gray value threshold and the defect detection model. The defect detection result includes the anomaly type of the photovoltaic solar cell corresponding to the newly added solar cell image.

[0023] In one embodiment, sending the image path to the worker process includes:

[0024] The main process writes the image path to the path queue and creates the corresponding worker process for the image path.

[0025] The main process sends the image paths in the path queue to the worker processes.

[0026] In one embodiment, the image path includes multiple image paths for newly added battery cell images, and the working process includes a working process corresponding to each image path; the main process sends the image paths in the path queue to the working processes, including:

[0027] The main process sends each image path in the path queue to the corresponding worker process.

[0028] Secondly, this application also provides a defect detection system, comprising:

[0029] Multiple testing devices are used to test the photovoltaic cells on the photovoltaic cell production line corresponding to the testing devices, obtain images of newly added cells, and store the images of newly added cells.

[0030] The inspection server communicates with multiple inspection devices and performs photovoltaic cell defect inspection methods as described in any of the first aspects.

[0031] Thirdly, this application also provides a photovoltaic cell defect detection device, comprising:

[0032] The acquisition module is used to obtain the image path of the newly added solar cell image stored on the detection equipment corresponding to each photovoltaic cell production line through the main process, and send the image path to the working process;

[0033] The segmentation module is used to obtain the image of the newly added battery cell corresponding to the image path through the working process, and to segment the image of the newly added battery cell based on the preset number of quadrants to obtain the image of each quadrant corresponding to the image of the newly added battery cell.

[0034] The detection module is used to perform brightness and darkness defect detection on the solar cell image based on at least one quadrant image to obtain the defect detection result of the photovoltaic solar cell.

[0035] Fourthly, this application also provides a detection server, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps provided in the first aspect.

[0036] Fifthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps provided in the first aspect.

[0037] Sixthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the method steps provided in the first aspect.

[0038] The aforementioned photovoltaic cell defect detection method and system acquire image paths of newly added cell images stored on the detection equipment corresponding to each photovoltaic cell production line through a main process, and send the image paths to a working process. The working process acquires the newly added cell images corresponding to the image paths, segments the newly added cell images based on a preset number of quadrants, and obtains quadrant images corresponding to each newly added cell image. Based on at least one quadrant image, brightness and darkness defect detection is performed on the cell image to obtain the photovoltaic cell defect detection result. In this embodiment, the detection server segments the newly added cell image to obtain quadrant images corresponding to each newly added cell image. Based on at least one quadrant image, brightness and darkness defect detection of the cell image does not require brightness and darkness defect detection of the entire newly added cell image to obtain the photovoltaic cell defect detection result, thus improving the defect detection efficiency of photovoltaic cells. Moreover, by collecting newly added cell images from all photovoltaic cell production lines onto a single detection server, the detection server can perform brightness and darkness defect detection on newly added cell images from multiple photovoltaic cell production lines, achieving one-to-N capability, not limited by the number of photovoltaic cell production lines, reducing the hardware deployment cost of brightness and darkness defect detection for photovoltaic cells, and reducing resource consumption. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of a defect detection system in one embodiment;

[0041] Figure 2 This is a flowchart illustrating a photovoltaic cell defect detection method in one embodiment;

[0042] Figure 3 This is a flowchart illustrating a method for determining defect detection results in one embodiment;

[0043] Figure 4 This is a flowchart illustrating a method for determining defect detection results in another embodiment;

[0044] Figure 5 This is a structural block diagram of a photovoltaic cell defect detection device in one embodiment;

[0045] Figure 6 This is a diagram of the internal structure of the detection server in one embodiment. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0048] The photovoltaic cell defect detection method provided in this application embodiment can be applied to, for example... Figure 1 The defect detection system shown includes multiple detection devices and a detection server. The detection server enables one-to-many detection, with one detection server inspecting multiple lines. Each detection device corresponds to one photovoltaic cell production line and performs individual detection on the photovoltaic cells on its respective production line.

[0049] The perception layer is equipped with multiple detection devices that inspect photovoltaic cells on the corresponding photovoltaic cell production line, obtaining images of newly added cells and storing them in the production line folder. The detection server comprises a data acquisition layer, an algorithm processing layer, and an application layer. The data acquisition layer uses a watchdog timer to monitor the cell images in the production line folder of the detection devices, obtaining the image path of the newly added cell image and storing it in a path queue. The algorithm processing layer creates multiple worker processes to retrieve image paths from the path queue, obtain the newly added cell images, perform initial screening of images in each quadrant of the newly added cell images, and perform fine screening using a defect detection model to obtain defect detection results. The application layer can perform production line anomaly analysis, missed detection sorting, and sensitivity configuration based on the defect detection results.

[0050] In one exemplary embodiment, such as Figure 2 As shown, a method for detecting defects in photovoltaic cells is provided, which can be applied to... Figure 1 The following explanation uses the detection server as an example, including the following steps S201 to S203. Wherein:

[0051] S201: The main process obtains the image path of the newly added solar cell image stored on the testing equipment corresponding to each photovoltaic cell production line, and sends the image path to the working process.

[0052] In this embodiment, each photovoltaic cell production line has a corresponding testing device. This device collects images of the cells from the corresponding photovoltaic cell production line and stores them in a production line folder. The main process monitors the production line folder on the testing device to determine if any new cell images are present. If a new cell image is found, the main process retrieves the image path of the new cell image stored on the testing device for each photovoltaic cell production line and sends this image path to the worker processes.

[0053] Optionally, the worker process can be created by the main process after obtaining the image path of the new battery cell image, or it can be pre-created. If the worker process is pre-created, after obtaining the image path of the new battery cell image, the main process determines one of the worker processes based on the working status of each worker process and sends the image path of the new battery cell image to that worker process. If the worker process is created after the main process obtains the image path of the new battery cell image, the image path of the new battery cell image can be placed in a queue, and after the corresponding worker process is created, the image path is sent from the queue to the worker process.

[0054] Optionally, a watchdog can be used to monitor and detect the production line folder on the device.

[0055] S202, the newly added battery cell image corresponding to the image path is obtained through the working process, and the newly added battery cell image is segmented based on the preset number of quadrants to obtain the quadrant images corresponding to the newly added battery cell image.

[0056] In this embodiment, the process acquires the image of the newly added battery cell corresponding to the image path, and segments the image of the newly added battery cell based on a preset number of quadrants to obtain images of each quadrant corresponding to the newly added battery cell image. For example, if the preset number of quadrants is 9, the image of the newly added battery cell is segmented in a 3×3 manner to obtain 9 quadrant images.

[0057] Optionally, the segmentation of the newly added battery cell image based on the preset number of quadrants can be done by equal division, meaning that the resulting images in each quadrant are of the same size. Alternatively, it can be done by arbitrary segmentation, meaning that the resulting images in each quadrant are of irregular size and / or shape.

[0058] S203, Based on at least one quadrant image, perform brightness and darkness defect detection on the solar cell image to obtain the defect detection result of the photovoltaic solar cell.

[0059] In this embodiment of the application, the photovoltaic cell defect detection result can be obtained by comparing the gray values ​​of at least one quadrant image with a preset gray threshold and performing brightness and darkness defect detection on the cell image based on the comparison result.

[0060] In one possible implementation, the texture and contour features of at least one quadrant image can be obtained, and the photovoltaic cell defect detection result can be obtained by performing brightness and darkness defect detection on the cell image based on the texture and contour features.

[0061] In another possible implementation, at least one quadrant image can be input into a preset detection model to obtain the defect detection results of the photovoltaic cell.

[0062] In the aforementioned photovoltaic cell defect detection method, the main process acquires the image paths of newly added cell images stored on the detection equipment corresponding to each photovoltaic cell production line, and sends the image paths to the working process. The working process acquires the newly added cell images corresponding to the image paths, segments the newly added cell images based on a preset number of quadrants, and obtains quadrant images corresponding to each newly added cell image. Based on at least one quadrant image, brightness and darkness defect detection is performed on the cell image to obtain the photovoltaic cell defect detection result. In this embodiment, the detection server segments the newly added cell image to obtain quadrant images corresponding to each newly added cell image. Based on at least one quadrant image, brightness and darkness defect detection of the cell image does not require brightness and darkness defect detection of the entire newly added cell image to obtain the photovoltaic cell defect detection result, thus improving the photovoltaic cell defect detection efficiency. Moreover, by collecting newly added cell images from all photovoltaic cell production lines onto a single detection server, the detection server can perform brightness and darkness defect detection on newly added cell images from multiple photovoltaic cell production lines, achieving one-to-N functionality, not limited by the number of photovoltaic cell production lines, reducing the hardware deployment cost of brightness and darkness defect detection for photovoltaic cells, and lowering resource consumption.

[0063] In one embodiment, the photovoltaic cell defect detection result is obtained by performing brightness and darkness defect detection on the cell image based on at least one quadrant image, including:

[0064] Based on the comparison between the gray values ​​of at least one quadrant image and a preset gray value threshold, the photovoltaic cell image is subjected to brightness and darkness defect detection to obtain the defect detection result of the photovoltaic cell.

[0065] Specifically, at least one quadrant image is converted into an array, and the array is converted into grayscale values ​​of the quadrant images.

[0066] In this embodiment, the process may involve sequentially determining the comparison results between the grayscale values ​​of each quadrant image and a preset grayscale threshold. If the grayscale value of the first quadrant image is greater than the preset grayscale threshold, then the comparison result between the grayscale value of the second quadrant image and the preset grayscale threshold is determined. If the grayscale value of the first quadrant image is not greater than the preset grayscale threshold, then the comparison result between the grayscale value of the second quadrant image and the preset grayscale threshold is not performed, thereby obtaining the defect detection result of the photovoltaic cell. For example, if the cell image includes four quadrant images, a first comparison result between the grayscale value of the first quadrant image in at least one quadrant image and the preset grayscale threshold is determined. If the first comparison result indicates that the grayscale value of the first quadrant image is greater than the preset grayscale threshold, a second comparison result between the grayscale value of the second quadrant image in at least one quadrant image and the preset grayscale threshold is determined. Finally, a comparison result between the second comparison result and the preset grayscale threshold is determined, thereby obtaining the defect detection result of the photovoltaic cell.

[0067] In another possible implementation, the process can simultaneously determine the comparison results between the gray values ​​of each quadrant image and a preset gray value threshold, and determine the defect detection result of the photovoltaic cell based on the comparison results. For example, the cell image includes four quadrant images, where the gray value of the first quadrant image is greater than the preset gray value threshold, the gray value of the second quadrant image is greater than the preset gray value threshold, the gray value of the third quadrant image is less than the preset gray value threshold, and the gray value of the fourth quadrant image is less than the preset gray value threshold. The defect detection result of the photovoltaic cell is determined based on the four comparison results.

[0068] In this embodiment, the photovoltaic cell defect detection result is obtained by comparing the gray values ​​of at least one quadrant image with a preset gray threshold to perform brightness and darkness defect detection on the cell image. This embodiment performs defect detection by comparing the gray values ​​of at least one quadrant image with a preset gray threshold. The defect detection method is simple and facilitates improved detection efficiency for photovoltaic cells, thereby increasing the manufacturing speed of photovoltaic cells.

[0069] Figure 3 This is a flowchart illustrating a method for determining defect detection results in one embodiment, as shown below. Figure 3 As shown, this application embodiment relates to a possible implementation method for obtaining the defect detection result of a photovoltaic cell by performing brightness and darkness defect detection on a cell image based on the comparison result of the gray value of at least one quadrant image with a preset gray value threshold, including the following steps:

[0070] S301, determine a first comparison result between the gray value of the first quadrant image in at least one quadrant image and a preset gray value threshold.

[0071] S302, if the first comparison result is that the gray value of the first quadrant image is greater than the preset gray value threshold, determine a second comparison result between the gray value of the second quadrant image in at least one quadrant image and the preset gray value threshold.

[0072] S303, if the second comparison result is that the gray value of the second quadrant image is not greater than the preset gray value threshold, the second quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell; the defect detection result includes the anomaly type of the photovoltaic cell corresponding to the newly added cell image, and the comparison result includes the first comparison result and the second comparison result.

[0073] S304, if the first comparison result is that the gray value of the first quadrant image is not greater than the preset gray value threshold, the first quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell.

[0074] In this embodiment, any one of the at least one quadrant images is taken as the first quadrant image. A first comparison result of the gray value of the first quadrant image and a preset gray value threshold is determined. If the first comparison result is that the gray value of the first quadrant image is greater than the preset gray value threshold, then any one of the at least one quadrant images other than the first quadrant image is taken as the second quadrant image. A second comparison result of the gray value of the second quadrant image and a preset gray value threshold is determined. If the second comparison result is that the gray value of the second quadrant image is greater than the preset gray value threshold, the comparison result of the gray value of the next quadrant image and the preset gray value threshold is determined. If the second comparison result is that the gray value of the second quadrant image is not greater than the preset gray value threshold, then the second quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell. In short, if the grayscale threshold of the previous quadrant image is greater than the preset grayscale threshold, the comparison between the grayscale threshold of the next quadrant image and the preset grayscale threshold continues. If there is a case where the grayscale threshold of the previous quadrant image is not greater than the preset grayscale threshold, the comparison between the grayscale threshold of the next quadrant image and the preset grayscale threshold is stopped. The quadrant images with grayscale thresholds not greater than the preset grayscale threshold are then input into the defect detection model to obtain the defect detection results of the photovoltaic cells. The defect detection results include the anomaly type of the photovoltaic cell corresponding to the newly added cell image. The anomaly type of the photovoltaic cell can be a known anomaly type or an unknown anomaly type. Cell images with unknown anomaly types can also be aggregated to optimize the defect detection model, laying the foundation for subsequent defect detection. For example, the time taken to detect 100 newly added cell images is reduced from 3.86 seconds to 0.39 seconds, an improvement of approximately 10 times in detection speed.

[0075] If the grayscale value of all images up to the last quadrant is greater than the preset grayscale threshold, then the defect detection result of the photovoltaic cell is determined to be that the cell is normal.

[0076] Optionally, when determining the comparison result between each quadrant image and the preset grayscale threshold, each grayscale value in the quadrant image can be compared with the preset grayscale threshold separately, or the maximum grayscale value in the quadrant image can be determined first, and then the maximum grayscale value can be compared with the preset grayscale threshold.

[0077] Optionally, grayscale values ​​of quadrant images can be randomly selected, and the first quadrant image, second quadrant image, etc., can be determined from each quadrant image using the randomly selected grayscale values.

[0078] Optionally, the defect detection model can be (You Only Look Once, YOLO).

[0079] Optionally, the grayscale values ​​of each quadrant image can be obtained before the grayscale comparison of the first quadrant image, or they can be obtained before the grayscale comparison of each first quadrant image. For example, if a new battery cell image includes four quadrant images, the grayscale values ​​of all four quadrant images can be determined in advance before determining the first comparison result between the grayscale value of the first quadrant image and a preset grayscale threshold. Alternatively, the grayscale value of the first quadrant image can be determined first, and if the grayscale value of the first quadrant image is greater than the preset grayscale threshold, the grayscale value of the second quadrant image can then be determined.

[0080] In this embodiment, a first comparison result is determined between the grayscale value of a first quadrant image and a preset grayscale threshold. If the first comparison result indicates that the grayscale value of the first quadrant image is greater than the preset grayscale threshold, a second comparison result is determined between the grayscale value of a second quadrant image and the preset grayscale threshold. If the second comparison result indicates that the grayscale value of the second quadrant image is not greater than the preset grayscale threshold, the second quadrant image is input into a defect detection model to obtain a defect detection result for the photovoltaic cell. Similarly, if the first comparison result indicates that the grayscale value of the first quadrant image is not greater than the preset grayscale threshold, the first quadrant image is input into the defect detection model to obtain a defect detection result for the photovoltaic cell. This embodiment, by sequentially performing defect detection on each quadrant image, eliminates the need to detect subsequent quadrant images if the grayscale value of a previous quadrant image is not greater than the preset grayscale threshold, thus reducing the computational burden on the detection server and improving the efficiency of defect detection. Furthermore, when the grayscale value of the quadrant image is not greater than the preset grayscale threshold, the defect detection model can be used to classify defects in photovoltaic cells, which facilitates the investigation of anomalies in the photovoltaic cell production line based on the defect detection results.

[0081] In one embodiment, such as Figure 4As shown, based on the comparison between the grayscale values ​​of at least one quadrant image and a preset grayscale threshold, the photovoltaic cell image is subjected to brightness and darkness defect detection to obtain the defect detection result of the photovoltaic cell, including:

[0082] Step A1: If the grayscale values ​​of all quadrant images corresponding to the newly added solar cell image are greater than the preset grayscale threshold, the defect detection result of the photovoltaic solar cell is determined to be that the photovoltaic solar cell corresponding to the newly added solar cell image is normal.

[0083] Step A2: If there is a gray value in each quadrant image corresponding to the newly added solar cell image that is not greater than a preset gray value threshold, determine the defect detection result of the photovoltaic solar cell based on the quadrant image that is not greater than the preset gray value threshold and the defect detection model. The defect detection result includes the anomaly type of the photovoltaic solar cell corresponding to the newly added solar cell image.

[0084] In this embodiment, the process simultaneously compares the grayscale values ​​of each quadrant image with a preset grayscale threshold. If the comparison result shows that the grayscale values ​​of all quadrant images are greater than the preset grayscale threshold, the defect detection result of the photovoltaic cell is determined to be that the photovoltaic cell corresponding to the newly added cell image is normal. If the comparison result shows that there is a quadrant image with a grayscale value not greater than the preset grayscale threshold, the quadrant image with a grayscale value not greater than the preset grayscale threshold is input into the defect detection model to obtain the defect detection result of the photovoltaic cell.

[0085] Optionally, all quadrant images not exceeding a preset grayscale threshold can be input into the defect detection model to obtain the defect detection results corresponding to each quadrant image not exceeding the preset grayscale threshold, and the defect detection results of the photovoltaic cell can be obtained based on the defect detection results corresponding to each quadrant image not exceeding the preset grayscale threshold; alternatively, a target quadrant image can be determined from multiple quadrant images not exceeding the preset grayscale threshold, and the target quadrant image can be input into the defect detection model to obtain the defect detection results of the photovoltaic cell.

[0086] In this embodiment, if the gray values ​​of all quadrant images corresponding to the newly added solar cell image are greater than a preset gray value threshold, the defect detection result of the solar cell is determined to be that the solar cell corresponding to the newly added solar cell image is normal. If there is a quadrant image corresponding to the newly added solar cell image whose gray value is not greater than the preset gray value threshold, the defect detection result of the solar cell is determined based on the quadrant image whose gray value is not greater than the preset gray value threshold and the defect detection model. This improves both the efficiency and accuracy of defect detection of solar cells.

[0087] In one embodiment, sending an image path to a worker process includes: writing the image path into a path queue through the main process and creating a worker process corresponding to the image path; and sending the image path in the path queue to the worker process through the main process.

[0088] This includes sending image paths from the path queue to worker processes via the main process; and sending each image path from the path queue to its corresponding worker process via the main process.

[0089] In this embodiment, the main process writes the image path of the newly added battery cell image into the path queue and creates a worker process corresponding to the image path. The main process then sends the image paths in the path queue to the worker processes. When there are multiple newly added battery cell images, multiple worker processes are started, and the image paths of the newly added battery cell images are sequentially placed into the path queue. The first-in-first-out (FIFO) principle is used to send each image path in the path queue to the corresponding worker process.

[0090] In this embodiment, the main process writes the image path into a path queue and creates a worker process corresponding to the image path; the main process then sends the image path in the path queue to the worker process. This embodiment uses a path queue to create worker processes and distribute image paths, thereby enabling rapid detection tasks and reducing hardware deployment costs.

[0091] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0092] Based on the same inventive concept, this application also provides a photovoltaic cell defect detection device for implementing the photovoltaic cell defect detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the photovoltaic cell defect detection device provided below can be found in the limitations of the photovoltaic cell defect detection method described above, and will not be repeated here.

[0093] In one exemplary embodiment, such as Figure 5As shown, a photovoltaic cell defect detection device is provided, including: an acquisition module 51, a segmentation module 52, and a detection module 53, wherein:

[0094] The acquisition module 51 is used to acquire the image path of the newly added solar cell image stored on the detection equipment corresponding to each photovoltaic cell production line through the main process, and send the image path to the working process;

[0095] The segmentation module 52 is used to obtain the newly added battery cell image corresponding to the image path through the working process, and to segment the newly added battery cell image based on the preset number of quadrants to obtain the quadrant images corresponding to the newly added battery cell image.

[0096] The detection module 53 is used to perform brightness and darkness defect detection on the solar cell image based on at least one quadrant image to obtain the defect detection result of the photovoltaic solar cell.

[0097] In one embodiment, the detection module 53 is specifically used to perform brightness and darkness defect detection on the solar cell image based on the comparison result of the gray value of at least one quadrant image with a preset gray value threshold to obtain the defect detection result of the photovoltaic solar cell.

[0098] In one embodiment, the detection module 53 is specifically used to determine a first comparison result between the gray value of the first quadrant image in at least one quadrant image and a preset gray value threshold.

[0099] If the first comparison result is that the gray value of the first quadrant image is greater than a preset gray value threshold, a second comparison result is determined between the gray value of the second quadrant image in at least one quadrant image and the preset gray value threshold.

[0100] If the grayscale value of the second quadrant image is not greater than the preset grayscale threshold in the second comparison result, the second quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell. The defect detection result includes the anomaly type of the photovoltaic cell corresponding to the newly added cell image, and the comparison result includes the first comparison result and the second comparison result.

[0101] In one embodiment, the detection module 53 is specifically used to input the first quadrant image into the defect detection model to obtain the defect detection result of the photovoltaic cell when the first comparison result is that the gray value of the first quadrant image is not greater than a preset gray value threshold.

[0102] In one embodiment, the detection module 53 is specifically used to convert each quadrant image in at least one quadrant image into an array, and to convert the array into grayscale values ​​of the quadrant images.

[0103] In one embodiment, the detection module 53 is specifically used to determine that the photovoltaic cell corresponding to the newly added cell image is normal when the gray values ​​of the images in each quadrant corresponding to the newly added cell image are all greater than a preset gray value threshold.

[0104] In one embodiment, the detection module 53 is specifically used to determine the defect detection result of the photovoltaic cell based on the quadrant image with a gray value not greater than a preset gray value threshold and the defect detection model when there is a gray value not greater than the preset gray value threshold in each quadrant image corresponding to the newly added cell image. The defect detection result includes the abnormal type of the photovoltaic cell corresponding to the newly added cell image.

[0105] In one embodiment, the acquisition module 51 is specifically used to write the image path into the path queue through the main process and create the working process corresponding to the image path; and to send the image path in the path queue to the working process through the main process.

[0106] In one embodiment, the acquisition module 51 is specifically used to send each image path in the path queue to the corresponding worker process through the main process.

[0107] Each module in the aforementioned photovoltaic cell defect detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the detection server in hardware form or independent of it, or stored in the memory of the detection server in software form, so that the processor can call and execute the corresponding operations of each module.

[0108] In one exemplary embodiment, a detection server is provided. This detection server may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the testing server includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores relevant data for photovoltaic cell defect detection. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a photovoltaic cell defect detection method.

[0109] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the solution of this application and does not constitute a limitation on the detection server to which the solution of this application is applied. A specific detection server may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0110] In one exemplary embodiment, a detection server is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above method embodiments.

[0111] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above method embodiments.

[0112] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of any of the above method embodiments.

[0113] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0114] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0116] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting defects in photovoltaic cells, characterized in that, The photovoltaic cell defect detection method includes: The main process obtains the image path of the newly added solar cell image stored on the testing equipment corresponding to each photovoltaic cell production line, writes the image path into the path queue, and creates a working process corresponding to the image path; the main process then sends the image path in the path queue to the working process. The newly added battery cell image corresponding to the image path is obtained through the working process, and the newly added battery cell image is segmented based on the preset number of quadrants to obtain the quadrant images corresponding to the newly added battery cell image. Determine the grayscale value of the first quadrant image in at least one of the quadrant images, and determine a first comparison result between the grayscale value of the first quadrant image and a preset grayscale threshold; If the first comparison result is that the gray value of the first quadrant image is greater than the preset gray value threshold, determine the gray value of the second quadrant image in at least one of the quadrant images, and determine the second comparison result between the gray value of the second quadrant image and the preset gray value threshold. If the second comparison result indicates that the grayscale value of the second quadrant image is not greater than the preset grayscale threshold, the second quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell; the defect detection result includes the anomaly type of the photovoltaic cell corresponding to the newly added cell image.

2. The photovoltaic cell defect detection method according to claim 1, characterized in that, The photovoltaic cell defect detection method includes: If the first comparison result indicates that the grayscale value of the first quadrant image is not greater than the preset grayscale threshold, the first quadrant image is input into the defect detection model to obtain the defect detection result of the photovoltaic cell.

3. The photovoltaic cell defect detection method according to claim 1, characterized in that, The photovoltaic cell defect detection method also includes: At least one quadrant image is converted into an array, and the array is converted into grayscale values ​​of the quadrant image.

4. The photovoltaic cell defect detection method according to claim 1, characterized in that, The image path includes multiple image paths for newly added battery cell images, and the working process includes a working process corresponding to each image path; the step of sending the image paths in the path queue to the working processes through the main process includes: The main process sends each image path in the path queue to the corresponding worker process.

5. A defect detection system, characterized in that, include: Multiple testing devices are used to test photovoltaic cells on the photovoltaic cell production line corresponding to the testing devices to obtain images of newly added cells, and the images of newly added cells are stored. The detection server is communicatively connected to the plurality of detection devices and executes the photovoltaic cell defect detection method as described in any one of claims 1-4.

Images