Surface defect detection and Neural Network recognition of automotive body panels

Abstract

In this new era, advances computer engineering in robotic vision to inspect the surface defect automatically is quickly penetrating this area of operation. The advance 3D scanning and image processing systems for quality are trending manufacturing industries. This paper proposes a solution for automated surface defect detection on automotive body panels in the context of quality control in industrial manufacturing. The 3D image is acquired from the 3D scanner from the surface of the body panel, then it's going through a process that will segment the uneven surface as potential defect area and by using Neural Network to recognize and classify defect area. The result of defect has been classified the system will show the defect occur on the automotive body panels.