Design And Implementation Of Workpiece Recognition And Positioning System Based On Shape Detection

With the implementation of the "Made in China 2025" strategy,China's manufacturing industry is getting higher and higher,and the application of industrial robots has become an important part of industrial production.The intelligent assembly production line needs to automatically identify and capture a large number of workpieces.Workpiece type identification and spatial attitude estimation are the key to accurate capture of the workpiece.In industrial production,there are many circular workpieces.Circular workpieces usually have no obvious texture features and no complicated shape features.It is difficult to estimate the pose by feature extraction.Aiming at this problem,this paper combines image processing technology and ellipse detection technology to estimate the pose of a brake disc with circular features.The specific application scenario is to estimate the pose of the brake disc in the material frame.The main work of this project is as follows:1.This paper briefly describes the research background,current situation and significance of workpiece identification and positioning,and introduces the related technologies of workpiece identification and positioning.2.A top-level workpiece recognition method based on edge segmentation is proposed.This method uses Harris corner detection algorithm to realize edge segmentation,combined with the contour and outer contour information of the workpiece,which can effectively identify the workpiece at the top layer.3.A method based on ellipse feature estimation is proposed.This method will form an ellipse(extremely a perfect circle or a straight line)in the image according to the circular target,combined with Canny edge detection algorithm,minimum The twomultiple ellipse fitting and the features of the workpiece itself realize the spatial pose estimation of the workpiece.This method can effectively eliminate the ambiguity of the spatial circular pose under the monocular camera.4.System design and implementation.Based on Windows,this paper designs a set of workpiece recognition and positioning system based on shape detection based on OpenCV.The system is based on MFC,using Visual Studio 2013 as a development tool,and the database is SQL Server 2008.The main functions are: user registration,login,top-level workpiece identification,workpiece pose estimation and other modules.