【摘要】 为提高液晶显示屏缺陷检测的速度和精度，设计了一种基于深度学习的液晶显示屏缺陷检测系统。针对实时检测系统中的Yolov5检测算法存在对全局信息的提取能力不足问题，在Transformer架构和C3模块基础上构建了C3TR模块并将其加入Yolov5基础模型。实验结果表明，所提出的算法在准确率和召回率上分别达到了90.9%和90.3%,与Yolov5基础算法相比分别提高了4.1%和1.4%。更多还原

【Abstract】 In order to improve the speed and accuracy of LCD defect detection, this paper designs an system for LCD defect detection based on deep learning. Aiming at the problem that the Yolov5 detection algorithm in the real-time detection system has insufficient ability to extract global information, this paper constructs the C3TR module based on the Transformer architecture and C3 module and adds it to the Yolov5 basic model. Experimental results show that the proposed algorithm reaches 90.9% and 90.3% in accuracy and recall, respectively, which is improved by 4.1% and 1.4% compared with Yolov5 basic algorithm, respectively.更多还原