基于YOLOv5的起重机制动下滑量测量系统设计  

Design of crane brake slip measurement system based on YOLOv5

作　　者：肖号迪 吴峰崎[2] 刘龙[1] 胡佳伟 Xiao Haodi;Wu Fengqi;Liu Long;Hu Jiawei

机构地区：[1]上海海事大学物流工程学院,上海200240 [2]上海市特种设备监督检验技术研究院,上海200062

出　　处：《起重运输机械》2024年第3期44-49,共6页Hoisting and Conveying Machinery

基　　金：国家市场监督管理总局科技计划项目“基于云模型的起重机械金属结构综合评估技术研究”(2022MK032);上海市市场监督管理局科技计划项目“基于视觉与位姿融合的动态测量系统关键技术研究”(2022-31)资助。

摘　　要：针对起重机制动下滑量测量问题,现阶段常见检测方法不能满足测量需求,故设计了一种基于YOLOv5的起重机制动下滑量检测软件。文中通过相机采集起重机制动过程,利用YOLOv5s训练并识别标记点,对标记点进行边缘提取和中心定位,改进欧式距离进行图像匹配,使匹配结果更加精确。实验结果表明,改进欧氏距离在匹配上更为精确,明显提高了起重机制动下滑量的测量精度,且下滑量测量系统具有高效性和稳定性。Currently,the common measurement methods of crane braking slip can not meet the measurement requirements.Therefore,in this paper,a crane braking slip detection software based on YOLOv5 is proposed.The braking process of crane was captured by camera,and the marked points were trained and identified by YOLOv5s,so as to extract the edges and locate the center of the marked points,and the Euclidean distance was modified for image matching,which made the matching result more accurate.The experimental results show that the modified Euclidean distance improves the matching accuracy,obviously increases the measurement accuracy of crane braking slip,and the slip measurement system is efficient and stable.

关 键 词：起重机 制动下滑量 YOLOv5目标检测 特征点匹配 检测软件 设计 

分 类 号：TH215[机械工程—机械制造及自动化]