You-Only-Look-Once Multiple-Strategy Printed Circuit Board
Defect Detection Model
Abstract
Printed Circuit Board (PCB) defect detection is crucial for ensuring the reliability and performance of
electronic devices. Traditional inspection methods are often labor-intensive and prone to error. This study
presents a novel approach utilizing the You Only Look Once (YOLO) object detection framework to
automate PCB defect detection. YOLO’s real-time object detection capabilities are harnessed to identify
and classify various types of defects on PCBs, such as soldering issues, component misalignments, and
surface anomalies. The proposed method involves training a YOLO model on a dataset of annotated PCB
images, followed by an evaluation of its performance in detecting defects across different PCB designs and
conditions. The results demonstrate that YOLO significantly improves detection accuracy and processing
speed compared to conventional methods. This approach not only enhances inspection efficiency but also
reduces the likelihood of human error, contributing to higher quality and reliability in electronic
manufacturing.
--Jishnujith k
S7EC2
TLY21EC042
