摘要
随着我国动车组保有量迅速增长,动车组检修工作压力日益增加。为了提高动车组运用检修效率,提高检准率,改善工人工作环境,本文设计了一种基于机器视觉和激光点云的动车组车顶螺栓松动检测装置。通过相机和雷达采集图像,经过预处理后通过Canny算子提取图像边缘特征线,分割出螺栓所在区域,再利用rgb2hsv算法,将图像转入HSV颜色空间,进而提取出防松标记线所在位置并进行拟合,判断螺栓是否出现松动。经测试,本系统精度可达0.25mm,检测准确率达到92.1%,符合现场工作要求。
With the rapid growth of the number of EMUs in China,the pressure of EMU maintenance work is increasing day by day.In order to improve the efficiency of EMU maintenance,improve the accuracy rate and improve the working environment of workers,this paper designed a loose detection device for roof bolts of EMUs based on machine vision and laser point cloud.The image is collected by the camera and radar,and after preprocessing,the image edge feature line is extracted by the Canny operator,the area where the bolt is located is segmented,and then the rgb2hsv algorithm is used to transfer the image to the HSV color space,and then the position of the anti-loosening mark line is extracted and fitted to determine whether the bolt is loose.After testing,the accuracy of the system can reach 0.25mm,and the detection accuracy rate reaches 92.1%,which meets the requirements of on-site work.
作者
房庭逸
林祥
张嘉麒
陈泽宇
Fang Tingyi;Lin Xiang;Zhang Jiaqi;Chen Zeyu
出处
《智慧工厂》
2024年第5期72-77,共6页
Smart Factory
关键词
机器视觉
车顶螺栓
松动检测
Machine Vision
Roof Bolts
Loosening Detection
