A stereo optical comparison method for detection of metallic surface defects based on machine vision and laser triangulation

Abstract

Quality of metallic surfaces is essential to maintain performance and longevity of industrial products. The surface defects of products like holes and cracks not only affect their appearances and performance, but also may even result in potential danger to human health in some cases. Traditional surface inspection methods for metallic surfaces rely on manual inspection methods by naked eyes or with the help of close circuit cameras, presenting the problems of low efficiency, low accuracy, high labour intensity due to monotonous work. Today manual inspection is not possible in many applications, therefore an automated method with a high accuracy is desirable. In this paper we propose a novel stereo optical comparison method to detect defects on metallic surface based on machine vision and laser triangulation.