Detection of a Surface Defect on an Engine Block Using Computer Vision

Abstract

Today the manufacturing world is facing major pressures due to the globalization of markets. Internal and external organizational pressures have led to increased competition, market complexity, and new customer demands. It has been noted that organizations and companies adopt lean or agile manufacturing strategies to overcome this problem. One of those strategies, the use of artificial intelligence and machine vision in the automotive industry, appeared as a good opportunity created by the continuous development of the computer-integrated manufacturing systems and the digitalization of the industry. In the automotive manufacturing process, developing a new automated defect detection system (ADDS), adapted for the automotive manufacturing requirements and particularities, and improved with artificial intelligence techniques, it is of major importance. In this paper we focus on a specific problem from the automotive industry, the problem of testing engine blocks using the information from real images.