A Machine Vision Approach for Detecting and Inspecting Circular Parts

In this paper, we present a machine vision approach for detecting and inspecting circular parts and parts with circular arcs on the contours. The method uses the Hough transform technique and uses the directional information of a normal to the circle at each boundary point. Cubic polynomial curve fitting is used to estimate the normal, and determine the concavity of the fitted curve at each given boundary point. The proposed Hough transform method is a two-stage pro-cedure. The first stage uses a 2D accumulator array to detect circle centres. The second stage uses a 1D accumulator array to detect the radii of circles. The proposed method is robust for detecting circular parts with partial occlusion such as peripheral defects or burrs. For an image of size N × N, the storage requirements are N 2 and the time complexity is bounded by (N+m)n, where m is the number of circle centres detected in the first stage and n is the number of boundary points in the image.