电磁干扰检测机器人的视觉引导系统  

Visual guidance system for electromagnetic interference detection robot

作　　者：刘鹏 王爱民[1] LIU Peng;WANG Aimin(School of Instrument Science and Engineering,Southeast University,Nanjing 210096,China)

机构地区：[1]东南大学仪器科学与工程学院,江苏南京210096

出　　处：《传感器与微系统》2025年第2期123-126,共4页Transducer and Microsystem Technologies

摘　　要：为实现电磁干扰(EMI)检测的自动化工作,设计了一种基于视觉图像的机器人检测引导系统。该系统通过使用双目视觉获取深度信息,使用Canny边缘检测并结合大津法来选取重要位置点,同时提高对视觉图像中纹理不明显处的检测能力。后续利用随机抽样一致性(RANSAC)算法纠正路径点偏差来获取一系列目标坐标值,控制机械臂携带近场探头进行检测扫描。使用UR3机械臂、D435摄像头以及近场磁场探头进行验证,能依据选取路径点测量较强磁场位置,得到磁场源位置信息。To achieve automation of electromagnetic interference(EMI)detection,a robot detection guidance system based on visual images is designed.This system obtain depths information by using binocular vision using.Canny edge detection algorithm combined with Otsu algorithm to select important location points and improve the detection ability for unclear texture areas in visual images.Subsequently,the random sample consensus(RANSAC)algorithm is used to correct path point deviations to obtain a series of target coordinate values.Finally,the mechanical arm carrying a near-field probe is controlled to carry out detection and scanning.UR3 robot,D435 camera and near-field magnetic field probe are used for verification.Strong magnetic field positions can be detected according to selected path points to obtain magnetic source position information.

关 键 词：双目视觉 轮廓检测 机械臂 电磁干扰 

分 类 号：TP242.2[自动化与计算机技术—检测技术与自动化装置]