光学精密工程2023，Vol.31Issue(22)：3357-3370,14.DOI:10.37188/OPE.20233122.3357
基于Swin Transformer轻量化的TFT-LCD面板缺陷分类算法
A lightweight deep learning model for TFT-LCD circuits defect classification based on swin transformer
摘要
Abstract
Defect detection in thin film transistor-liquid crystal display(TFT-LCD)circuits is a challeng-ing task because of the complex background setting,different types of defects involved,and real-time de-tection requirements from industry.Traditional methods have difficulties in satisfying the dual require-ments of detection speed and accuracy.To address this challenge,in this study,a deep learning method is developed for image classification based on the Swin Transformer technique.First,token merging is used to reduce the computational complexity of each layer of the model,thus improving computation efficiency.Then,a depthwise separable convolution module is introduced to add convolutional bias to reduce the reli-ance on massive data.Finally,a knowledge distillation method is applied to overcome the problem of re-duced detection accuracy caused by the less-intensive computation design.Experimental results on the self-made dataset demonstrate that the proposed method achieves a 2.6 G FLOPs reduction and a 17%speed improvement compared to baseline models,with only a 1.3%Top-1 accuracy precision reduction.More importantly,the proposed model achieves better balance on accuracy and detection speed on both self-made and public datasets than existing mainstream models on image classification in the TFT-LCD manu-facturing industry.关键词
TFT-LCD/Transformer/图像分类/计算机视觉Key words
Thin Film Transistor Liquid Crystal Display(TFT-LCD)/transformer/image classifica-tion/computer vision分类
信息技术与安全科学引用本文复制引用
夏衍,罗晨,周怡君,贾磊..基于Swin Transformer轻量化的TFT-LCD面板缺陷分类算法[J].光学精密工程,2023,31(22):3357-3370,14.基金项目
国家自然科学基金资助项目(No.51975119) （No.51975119）
无锡市"太湖之光"科技攻关(产业前瞻及关键技术研发)项目资助(No.G20222011) （产业前瞻及关键技术研发）
