A Neuro-Fuzzy Approach to a Machine Vision-based Parts Inspection Problem

This paper describes a research project whose objective is to improve the parts inspection component of the automotive manufacturing process through the application of neuro-fuzzy systems. The basic methodology is to circulate case studies of industrial inspection problems among 5 universities and challenge the researchers to find more robust analysis algorithms. This paper presents initial work on one of the case studies whose subject is the application of a machine vision-based system to identify missing fasteners in a cross-car beam.