| Machine vision is gradually improved and perfected with the development of the Industrial Automation.It plays a vital role in effective and high quality production.productive efficiency and labour condition can be improved with useing Machine vision technic on welding robot,it can even achieve unmanned welding, effectively drive improvement of world production techniques.Based on the the rsearch of curve rcognition arithmetic, the author designs a 2D curve detection and tracking system. simplifys camera model and provides a method 3D curve rcognition arithmetic2D curve detection and tracking system is based on the developed tool of VC++ and adopts Open Source Computer Vision Library OpenCv, from which the image information is collected. After dilation and thinning and getting the bar frame, nearly single pixel wide curve central line are obtained.Then get the precise curve central line data with 2D curve seeking arithmetic.With the aid of the data, the robot of laboratory can be controlled by making it follow the locus of the curve. It can also achieves simulated running function.Research 3D curve recognition technology which is based on 2D curve recognition technology, predigest the camera model, restrict the place of the camera, using the relationship of the camera coordinate. Educe a 3D curve recognition arithmetic which is based on 2D curve recognition arithmetic, and analyzed the concrete realization step of the arithmetic.Introduce how to develope two camera using program with DirectShow, Lay the groundwork of 3D curve recognition program development.2D curve detection and tracking system integrate the machine vision and the laboratory equipment, increases the content and new ground of the robot laboratory.the image processing method can be used on industry welding robot, achieve self welding. It has a certain value in the application of industry application.3D curve recognition arithmetic which supplys a new idea in the Research of 3D vision, is workable in theory. |
PDF Full Text Request