光学精密工程2025，Vol.33Issue(9)：1434-1445,12.DOI:10.37188/OPE.20253309.1434
Micro LED表面缺陷的快速高精度检测
Rapid and high-precision detection on surface defects of Micro LED
摘要
Abstract
To address the demands for real-time and high-precision Micro LED defect detection,this study introduces LED-YOLO,a rapid and accurate detection algorithm that integrates a lightweight archi-tecture with enhanced feature extraction capabilities.An image acquisition system was designed to simu-late industrial interference,and various data augmentation techniques were employed to increase the diver-sity of training data.To overcome the limited discriminative power for Micro LED defects,a Lightweight Dynamic Fusion Module(LDFM)was developed,combining dynamic convolution,deep convolution,and channel mixing operations;this approach maintains model compactness while enhancing feature extrac-tion.Furthermore,an Enhanced Coordinated Attention Module(ECAM)was proposed to improve de-fect localization by integrating channel and spatial attention mechanisms alongside residual connections,thus refining feature extraction accuracy.Given the minimal aspect ratio variation in Micro LED images,a dynamic focusing mechanism was incorporated,and a DIoU_W regression loss function was introduced to accelerate convergence and improve robustness.Experimental results demonstrate that LED-YOLO sur-passes the state-of-the-art YOLOv11s in detection accuracy,recall,mean average precision(mAP),and F1 score.Despite a reduction of 1.6 million parameters,LED-YOLO achieves substantial improvements in detection speed and accuracy,effectively fulfilling the quality inspection requirements of Micro LED panel manufacturing.关键词
深度学习/Micro LED/缺陷检测/动态卷积/注意力机制Key words
deep learning/Micro LED/defect detection/dynamic convolution/attention分类
信息技术与安全科学引用本文复制引用
赵天元,董登峰,王国名,王博,周维虎..Micro LED表面缺陷的快速高精度检测[J].光学精密工程,2025,33(9):1434-1445,12.基金项目
国家重点研发计划资助项目(No.2021YFF0700304) （No.2021YFF0700304）
