基于视觉的汽车线束绑扎胶套检测与测量系统  

Detection and Measurement System of Automobile Wire Harness Binding Rubber Sleeve Based on Vision

作　　者：马国庆[1,2] 刘珺玮 曹国华 MA Guo-qing;LIU Jun-wei;CAO Guo-hua(School of Mechanical and Electrical Engineering,Changchun University of Science and Technology,Changchun 130022,China;Wuhu Hit Robot Technology Research Institute Co.,Ltd,Wuhu 241000,China;Chongqing Research Institute of Changchun University of Science and Technology,Chongqing 401133,China)

机构地区：[1]长春理工大学机电工程学院,吉林长春130022 [2]芜湖哈特机器人产业技术研究院有限公司,安徽芜湖241000 [3]长春理工大学重庆研究院,重庆401133

出　　处：《仪表技术与传感器》2022年第9期82-88,共7页Instrument Technique and Sensor

基　　金：中国博士后科学基金(2021M692457);吉林省科技发展计划(20190302122GX);吉林省教育厅科学研究规划项目(JJKH20200747KJ);重庆市自然科学基金(cstc2021jcyj-msxmX0989)。

摘　　要：为解决汽车电路线束绑扎胶套的质量检测时人工测量误差大、效率低的问题,提出基于机器视觉和深度学习的汽车线束绑扎胶套检测与测量系统。系统通过相机对汽车线束的各个部分进行图像采集,然后利用Halcon算法库对采集的多张图像进行拼接,通过深度学习和形态学处理对拼接图像中的胶套进行粗定位和精定位确定胶套数量及位置,使用自适应边缘检测阈值设定法检测胶套边缘,提出一种自适应类高斯权重分布的全采样测量方法对胶套长度进行测量。经现场测试,该系统对胶套尺寸的测量平均相对误差低至1.12%,胶套检出率达100%,可满足实际应用中胶套检测与测量要求。In order to solve the problems of large manual measurement error and low efficiency in the quality inspection of automotive wiring harness binding rubber sleeves,a detection and measurement system for automotive wiring harness binding rubber sleeves based on machine vision and deep learning was proposed.The system collected images of various parts of the car wiring harness through the camera,and then used the Halcon algorithm library to stitch the collected images,and used deep learning and morphological processing to roughly locate and finely locate the rubber sleeves in the spliced images to determine the number and position of rubber sleeves.The edge of the rubber sleeve was detected by the adaptive edge detection threshold setting method,and a full-sampling measurement method with an adaptive Gaussian-like weight distribution was proposed to measure the length of the rubber sleeve.After on-site testing,the system has an average relative error of as low as 1.12%in the measurement of the size of the rubber sleeve,and the detection rate of the rubber sleeve is 100%,which can meet the requirements of rubber sleeve detection and measurement in practical applications.

关 键 词：机器视觉 深度学习 图像拼接 边缘检测 权重分布 全采样测量 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]