Research On Target Recognition And Location Of Industrial Robot Based On Computer Vision

With the rapid development of computer,image processing and image recognition technology,machine vision is becoming more and more common.Machine vision is one of the main means for industrial robots to obtain environmental information.Improve robot autonomy and flexibility.It is widely used in manufacturing and other industries to ensure product quality,control production processes and sense the environment.Industry 4.0 is inseparable from intelligent manufacturing,which is inseparable from machine vision.Industrial robots can extract functional parameters and position information of the workpiece through visual image information,identify the type of the workpiece and determine the position of the workpiece.Machine vision is a major technology applicable to industry.Therefore,taking hsr-612 industrial robot as an example,the target recognition and position algorithm of machine vision are studied.The real-time matchingrecognition technology and spatial location technology are studied in detail.The main tasks are as follows:(1)in order to improve the quality of the images of the assembled workpiece obtained from the CCD camera,the captured images are preprocessed mainly including histogram equalization and median filtering.Secondly,the edge characteristics of the samples were extracted,and through the analysis and comparison of various classical edge detection operators,the accurate positioning of the canny operator was obtained,which had a single response and was able to obtain a good edge detection effect.(2)based on existing image matching algorithms,a workpiece recognition algorithm based on Hausdorff distance and an improved genetic algorithm were proposed.The algorithm USES the edge feature of the workpiece as the scale of the contour similarity of the object,improves the matching feature and corrects the hausdorff distance.The improved genetic algorithm is used to search the target quickly and achieve the best match.In the space of distance conversion,the target object has been successfully transmitted.Experimental results show that the proposed recognition algorithm can effectively detect changes and occlusion in a small range of translation and rotation.(3)when determining the position of the working space of the moving target,based on a certain rotation matrix,the moving view of a single object is used to obtain the depth information of the processed object and process itinto 3d.By keeping the rotation matrix of axis J 4-j 6 relative to the basic coordinate system of the industrial robot,the spatial coordinates can be obtained directly.(4)take hsr-612 industrial robot as the execution matrix,take the industrial robot technology application skills competition platform as the research platform,use CCD camera,image acquisition card and PLC controller to build the robot human vision system,use the experimental device,use the algorithm of identifying and locating the work piece,so as to obtain and assemble the work piece.