Research on Defect Detection Method of Printed Circuit Board Based on Improved Yolov8
In order to solve the problem of low detection accuracy and low detection efficiency of small target defects on the surface of the Printed Circuit Board,this paper proposes the Printed Circuit Board defect detection method based on improved YOLOv8.The Simplified Spatial Pyramid Pooling Fast module is used in the YOLOv8 model backbone network,and the Rectified Linear Unit activation function is used to effectively improve the detection efficiency.Using the improved Bottleneck Attention Module,the channel attention branch and the spatial attention branch are used to make the model have a strong representation ability to obtain feature information and improve the detection accuracy.The loss function is improved,and the WIoU loss function is constructed to reduce the influence of geometric factors when the anchor box coincides with the target box.On the defect data set of Printed Circuit Board,the detection accuracy mAP@0.5 value of the network model reaches 90.2%,which is 5.7%higher than that of the original network model,which is of great significance to the defect detection of Printed Circuit Board.
国家科技期刊平台
NSTL
万方数据
