Detecting Occlusions of Automobile Parts Being Inspected By a Camera System During Manufacturing Assembly

This thesis considers the problem of detecting occlusions in automobile parts on a moving assembly line in an automotive manufacturing plant. This work builds on the existing “Visual Inspector” (VI) system developed as a joint research project between Clemson University and the BMW Spartanburg manufacturing plant. The goal is to develop a method that can successfully detect occlusions in real-time. VI is a detector and classifier system that uses video cameras to determine the correct installation of a part in the assembly line. In the current version of VI, an occluded part is flagged simply as ‘not OK’ as if the part were not installed at all. The new algorithm developed aims to extend the functionality of VI to correctly identify occlusions i.e., flag an obscured, but correctly-installed part as ‘occluded’ rather than as ‘not OK’. In this thesis, we provide a background of the current VI system deployed at the manufacturing plant. We then discuss the design of an algorithm that recognizes occlusions. Details of tests conducted to verify the correctness of the design, as well as the results of the tests run on real-world data from the plant are presented. Finally, we discuss the possible enhancements to this algorithm as part of future work.