Review on application research of robots scraping system based on visual

XIA Qun-feng，PENG Yong-gang

（College of Electrical Engineering，Zhejiang University，Hangzhou 310027，China）

Abstract：Aiming at the problem that the level of intelligence and flexibility of industrial robots in production line is not high，computer vision technology which is advancing rapidly were introduced to the original field of industrial transfer robot. Machine vision technology was used to obtain information on the workpiece and its surroundings and target artifact to be operated was identified. Decisions were made by using of this technology to guide industrial robot to complete some operations on artifact，such as picking and placing. Three aspects of the research result at present stage were reviewed，which was aiming at industrial robot scraping system in production line， such as camera calibration，identification and matching of target artifact，positioning and scraping to artifact by robot. The image preprocessing methods which were involved in computer vision were analyzed and summarized. The practical application and future development of the technique were discussed. The results indicate that this scraping system has strong technology supported，and it can satisfy the real-time requirement of industrial application. Each part of the research and improvement of the algorithm have a certain reference value for industrial development and related studies.