A deep learning method is proposed in this paper to combine YOLOv5 into solar panel defect detection and optimize it.Firstly,coordinate attention mechanism is introduced to enhance the target feature,improve the capture degree of feature information by YOLOv5 neck network,and enable the whole network to locate and identify the target area more accurately.The C3 module is improved to the CoT module to improve the network to make full use of the context informa-tion near the defect features of solar panels and accelerate the convergence speed.The accuracy rate and recall rate of the improved model on the test set reached 92.7％and 93.6％respectively,and the average precision average(mAP)reached 95％,which was 2％higher than that of the original network.
关键词
深度学习/缺陷检测/太阳能电池板/YOLOv5
Key words
deep learning/defect detection/solar panels/YOLOv5
维普
万方数据