【摘要】 基于现有表面缺陷检测系统所存在的实时检测难、硬件要求高等问题，提出一种基于云计算与边缘协同计算的表面缺陷检测系统。将轻量化改进后的YOLOv4缺陷检测算法模型部署到边缘端嵌入式设备中，在边缘端完成对表面缺陷的检测，并在边缘端和云端设备部署KubeEdge框架进行通信和管理。通过案例验证该系统不仅能够满足检测实时性的要求，还能够提取缺陷检测关键信息，同时便于部署在价格低廉的嵌入式设备。更多还原

【Abstract】 Existing surface defect detection system has a series of problems such as the difficulty of real-timedetection and the high hardware requirements.To solve these problems, a surface defect detection system is proposed using the cloud-edge computing.The system deploys the lightweight YOLOv4 defect detection algorithm model to the edge-end embedded device, completes the detection of surface defects at the edge-end, and deploys the KubeEdge framework on the edge and cloud devices for communication and management.Through case verification, the proposed system can not only meet the requirements of real-time detection, but also extract key information about defect detection, and is easy to deploy in low-cost embedded devices.更多还原