Research on edge detection of thin sheet part dimension inspection system based on machine vision
Aiming at the edge detection of thin sheet part dimension inspection system based on machine vision, a contrast research on edge detection is investigated. The Gaussian blurred simulation image and thin sheet part image are took as evaluation images, and the edge detection are done with Roberts operator, Sobel operator, Prewitt operator, Kirsch operator, Laplacian operator, LOG operator and mathematical morphology edge detection method. The results of edge detection are analyzed deeply, and the edge location accuracy, noise resisting ability and calculation time of each algorithm are compared. The single-pixel width connected contour is acquired with mathematical morphology edge detection method, the detection time are 0.0521 second and 0.457 second respectively. It is appropriate that taking the mathematical morphology edge detection method as the edge detection method of thin sheet part dimension inspection system based on machine vision.
machine vision dimension inspection system edge detection location accuracy noise resisting ability calculation time
Wu JiGang He Kuanfang Qin Bing
Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment, Hunan University of Engineering Research Center of Advanced Mine Equipment, Ministry of Education, Hunan University of S
国际会议
湘潭
英文
194-198
2011-07-19（万方平台首次上网日期，不代表论文的发表时间）