基于HALCON的单目视觉工件位姿测量方法研究  

Method Study on Position and Orientation Measurement of Monocular Vision Workpiece Based on HALCON

作　　者：周甫[1] 郑堤[2] 张松[1] 

机构地区：[1]宁波大学机械工程与力学学院,浙江宁波315211 [2]浙江大学宁波理工学院机电与能源工程学院,浙江宁波315100

出　　处：《机械制造》2017年第2期85-90,共6页Machinery

摘　　要：为了满足单目视觉系统对被测工件三维位姿参数测量及机器人实时快速响应要求,提出一种通过单目视觉获取规格已知、位置随机工件三维参数的方法。利用HALCON图像处理软件对机器人和工业相机进行系统标定,通过标定参数进行采集图像预处理、点特征坐标提取,引入约束条件建立三维位姿算法几何模型,完成上位机图像采集与数据处理,并实时将数据通过以太网传送至机器人,实现工件抓取。试验表明,这一方法误差范围达到了工业机器人操作的预期要求。In order to meet the fast response requirement when using monocular vision system and robot to measure 3D position and orientation parameters of the workpiece in real-time, a method was proposed to obtain the 3D parameters of the workpiece with known position and random position by monocular vision. By using the HALCON image processing software, the robot and the industrial camera were systematically calibrated. The image preprocessing was carried out by the calibrated parameters. The point characteristic coordinates were extracted and the geometric model of the 3D position and orientation algorithm was established by introducing the constraint condition. The image acquisition and data processing were achieved by the upper computer, and real-time data would be sent to the robot via Ethernet to achieve the pick-up of the workpiece. The experimental results show that the error range of this method can meet the expectation of industrial robots.

关 键 词：单目视觉 位姿 计算机 测试 

分 类 号：TH122[机械工程—机械设计及理论] TP242.2[自动化与计算机技术—检测技术与自动化装置]