Fast Measurement Method of Gear Parameters Based on Convexity Defect

Abstract

Gear parameters measurement based on machine vision is an important part in the field of intelligent manufacturing. Existing methods, which are all based on gear center, may cause accumulative error and have the weakness of large computation and low efficiency due to the repeated use of all contour points. In view of these problems, a new measurement method of gear parameters based on convexity defect is presented. Firstly, a gear contour is extracted from the gear image. Secondly, convex hull and defects of the gear contour are computed using image processing approach, then the number of the gear teeth and key points on the dedendum circle and addendum circle are determined, respectively. Finally, each radius of the dedendum circle and addendum circle is obtained separately by means of circle fitting with the corresponding key points. In this method, the key points are calculated by using the properties of convex defects, avoiding the calculation of the gear center and greatly increasing the computing speed by using a limited number of key points, precisely, two times the number of gear teeth. Compared with the traditional methods, the experimental results show that the speed of this method is increased by 1.5 times and it can meet the real-time requirement.