A Method for Surface Defect Detection of Printed Circuit Board Based on Improved YOLOv4

Abstract

The quality detection of printed circuit board (PCB, Printed Circuit Board) is a very important task. However, due to the diversity, complexity and smallness of PCB defects, traditional surface defect detection methods are still difficult to deal with. Therefore, this paper proposed a method for detecting surface defects of printed circuit board based on improved YOLOv4. Firstly, we used the multistage residual hybrid attention module (MRHAM) to improve feature learning to enhance the feature expression ability of the shallow network, so that the receptive field pays more attention to the object feature and ignores irrelevant features; Then, we used the K-means++ clustering algorithm perform cluster analysis on the experimental dataset, determined the anchor boxes value of the PCB defects dataset, and improved the accuracy of the model for small object defect positioning; Finally, we used online and offline data augmentation, transfer learning and multi-scale training methods to enhance the adaptability of the model to images of different input scales and improved the generalization ability of the model. Experimental results showed that compared with other traditional and deep learning object detection models, the improved model has improved detection accuracy and detection rate. The mAP value reaches 99.71% and the detection rate reaches 68 f.s <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-1</sup> . It showed that the improved YOLOv4 has high application value in PCB surface defect detection.