Elliptic Feature Recognition and Positioning Method for Disc Parts

Abstract

Machine vision technology is more and more widely used in industrial production. Parts’ sorting is a very common application scenario in industrial production. Considering the particularity of circular Parts, a method of part identification and positioning based on ellipticity is proposed. The method firstly preprocesses the image, then performs threshold segmentation to extract the edge contour, and then further filters the extracted edge contour, and performs least squares circle fitting on the basis of the retained edge contour to solve the Parts’ pose. Finally, similarity matching is performed to determine the type of Parts’ information. The experimental results show that the proposed algorithm can effectively identify and locate the parts.