Research On Industrial Robots Scraping Technologies Based On Machine Vision

Most of the handling industrial robots currently working on the production line are operated by pre-programming or utlize teaching,so that the initial and end position of the target object are strictly limited,and the flexible production system can not meet the requirements of the material conveying system.The purpose of this paper is that guide the machine robots to obtain the environmental information,and identify and locate the position of the target workpiece in real time,It is one of the key technologies of machine vision application in the industrial field.It can break the limitations of the industria robots,and improve the Versatility and intelligence of the robot.Therefore,this paper proposes a solution based on machine vision collaborative industrial robot system to locate and grab the workpiece.It mainly focuses on detailed research of the workpiece's real-time matching and spatial positioning technology,and the main contents are as follows:Firstly,the kinematics analysis of industrial robots is carried out in this paper.The kinematics positive solutions and the inverse solutions of robots are obtained by D-H coordinate transformation algorithm and the inverse transformation methods.Then,Zhang Zhengyou calibration method was used to calibrate the camera.The internal and external parameters of the camera and the transformation matrix were deduced,so we can establisht that the conversion relationship between the robot and the camera.Secondly,the preprocessing of the images include the image histogram equali-zation and median filtering;then extracting the edge features of the object,after analysis of each the edge detection means,it proves that using Canny detection method to obtain the good Edge features.Finally,based on the traditional image matching algorithm,this paper proposes an improved Hausdorff distance matching algorithm based on the image pyramid structure,we calculates the centroid position of the target,and then extrapolate the target of workpiece's Space position in the world coordinate system through the coordinate conversion matrix.So as to guide the industrial robot autonomously complete the accurate capturing of the target workpiece.According to the experimental results,the feasibility of the algorithm in this study has achieved the purpose of the robot to accurately grasp the workpiece.