摘要
机器人关节工件组装生产过程中,工件种类多、产量大、人工分拣与装配困难等问题,为了提高生产速度和效率,提出一种基于OpenCV视觉技术的机器人关节工件识别与定位检测方法。首先,通过工控机读取相机的工件图像,并进行图像灰度化、均值滤波、自适应阀值分割等图像预处理过程,得到二值化图像;其次,从二值化图像中提取出工件的轮廓,之后,通过胡氏不变矩比较待识别工件的轮廓与给定的模板轮廓,识别目标工件;最后,提出一种确定工件的位置和方向的方法。实验结果表明,文章提出的方法能准确识别目标工件,并且有效地确定其位置,为类似的识别提供了有效参考。
In the assembly process of robot joint workpiece,there are many kinds of workpiece,large output,manual sorting and assembly difficulties,etc.in order to improve the production speed and efficiency,a method of robot joint workpiece recognition and positioning detection based on OpenCV vision technology is proposed.Firstly,the image of the camera′s workpiece is read by the industrial computer,and the image pre-processing process,such as image graying,mean filtering,adaptive threshold segmentation,is carried out to get the binary image;Secondly,the contour of the workpiece is extracted from the binary image,and then the target workpiece is identified by comparing the contour of the workpiece to be identified with the given template contour through Hu′s moment invariants.Finally,a method to determine the location and direction of the workpiece is proposed.The experimental results show that the method proposed in this paper can accurately identify the target workpiece,and effectively determine its location,which provides an effective reference for similar identification.
作者
罗辉
崔亚飞
LUO Hui;CUI Ya-fei(School of Intelligent Manufacturing and Architectural Engineering,Yongzhou Vocational Technical College,Yongzhou Hunan 425100,China)
出处
《组合机床与自动化加工技术》
北大核心
2021年第1期101-105,109,共6页
Modular Machine Tool & Automatic Manufacturing Technique
基金
湖南省教育厅科学研究项目(18C1761)。
