Статьи

Solar cells are very prone to scratches, hot spots, breakage and other defects during the production process, which seriously affects their service life and photoelectric conversion efficiency. Traditional detection methods cannot meet the accuracy and real-time requirements of the actual industrial production. To address the problems of low detection accuracy, slow speed, and single type of detected defects in solar cell defect detection, this paper proposes a solar cell defect detection algorithm based on improved YOLOv8s, which is based on the original YOLOv8s network model, and introduces the GAM global attention mechanism module and the EIoU-Focal loss function. The experimental results show that compared with other algorithms, the mAP@0.5 of the improved YOLOv8s reaches 85.1%, and the algorithm has a better improvement in detection accuracy and detection effect, which can complete the task of detecting defects in solar cells more quickly and accurately.