Automatic Detection System for Defects on the Surface of Conelet Tapered Roller

Abstract

Aiming at the current problems of low efficiency and high cost when manually detecting the surface flaw of tapered roller, the automatic detection system based on machine vision is developed. Combining with the features of roller detection, the principle of automatic detection and work process are introduced. The design schemes such as automatic feeding device, unfolding device of Roller and automatic sorting device are proposed. In order to obtain clear and stable roller surface image, appropriate CCD camera and optical microscope are chose and at the same, the quality of lighting source shall be improved. The experiment shows that this program can achieve the desired results and realize real-time online detection for defects on the surface of roller in the state of motion.