YOLO V8 based Printed Circuit Board Fault Detection

Abstract

Printed Circuit Boards (PCBs) are important modules in numerous electronic maneuvers. Faults in PCBs could lead to malfunctioning or even failure of these devices. Automated defect detection is crucial to ensure high-quality PCB production. In this article, we explore the optimization of performance parameters for the YOLO v8 object detection model to improve PCB fault detection accuracy. PCBs are used in numerous electronic devices, like smartphones, medical equipment, etc. The presence of defects in PCBs could lead to malfunctioning or even failure of these devices. Therefore, it is essential to detect and correct these defects during the manufacturing process. Traditional visual inspection methods are time-consuming and horizontal to human error. Automated defect detection using computer vision algorithms could significouldtly improve the efficiency and accuracy of PCB quality control. YOLO is a general object detection procedure that has revealed promising fallouts in numerous applications, including PCB defect detection. The modern form, YOLO v8, offers improved accuracy and speed compared to its predecessors. In this article, we discussed the optimization of performance parameters for YOLO v8 to enhance its capability in detecting PCB faults. We evaluated the model’s enactment by metrics like precision, recall, and F1-score. The F1-Score is achieving upto 96%, Precision upto 100% and recall of 98%.