Optimum Illumination for Real-time Wiper Arm Defect Detection

Abstract

The importance of optimum illumination has been undervalued in machine vision system over the years. Most of the research work emphasized on developing advanced and complex algorithms with little attention on lighting design and configurations. Using the right lighting configuration can significantly improve the efficiency and effectiveness of an inspection system. This paper describes the design of a novel illumination system that reduces the complexity of the image processing algorithm in wiper arm surface defect detection. The lighting system increases the contrast between flawless area and defective area in the image captured. Thus, it facilitates a more accurate, reliable and productive machine vision system. The processing time has improved to 0.21 seconds per wiper arm.