Research Article

An Improved YOLOv5 Network for Detection of Printed Circuit Board Defects

Table 3

Comparison of methods on the PCB defect dataset.

Method[email protected] (%)[email protected]:.95 (%)Speed (FPS)Parameters (MB)

Faster R-CNN95.3264.57528
SDD84.6755.741149
RetinaNet89.761.223157
YOLOv497.6363.362248
YOLOv5s98.7266.18714.5
Proposed method99.169.38617