Bracket move inspection using canny edge detection

Abstract

A bracket, the plastic part attached to a pivot arm, is a significant part in the production of pivot arm.
Due to the small size of the component, it is difficult and exhausting to manually detect the error in attaching
process of the bracket. The goal of this research is to facilitate the process of quality control in detecting
bracket move in a pivot arm. In particular, we utilize the Canny edge detection in microscopic images of pivot
arms. We implement this method in a real-world data set. The result shows that we can achieve the estimated
distances between brackets and bore holes that are very close to the measurements by a smart scope.