Research and Development of Target Recognition and Location Crawling Platform based on Binocular Vision
The application of visual recognition technology in industrial robot crawling and placing operation is one of the key tasks in the field of robot research.In order to improve the efficiency and intelligence of the material sorting in the production line,especially to realize the sorting of the scattered items,the robot target recognition and positioning crawling platform based on binocular vision is researched and developed.The images were collected by binocular camera,and the images were pretreated.Harris operator was used to identify the corners of the images.The Canny operator was used to identify the images.Hough-chain code recognition was used to identify the images.The target image in the image,obtain the coordinates of each vertex of the image,calculate the spatial position and posture of the target item,and determine the information needed to capture the movement and transmit it to the robot control crawling operation.Finally,In this paper,we use this method to experiment the wrapping problem in the express sorting process The experimental results show that the platform can effectively solve the problem of sorting of loose parts,so as to achieve the purpose of efficient and intelligent sorting.
Weidong Xu Zhu Lei Zhang Yuan Zhenqing Gao
School of Mechanical Engineering,Beijing Institute of Graphic Communication,Beijing,China
国际会议
上海
英文
1-8
2017-12-28(万方平台首次上网日期,不代表论文的发表时间)