Automated visual inspection for surface profile tolerancing

Achieving inspection automation is desirable in a production line in order to reduce labour cost and ensure consistent product quality. This paper presents an automated inspection system for a three-dimensional complex surface according to the specification of surface profile tolerancing. A range scanner is used to measure the surface profile of an object and an algorithm is designed to decide upon the acceptability of the part. The pyramidal generalized Hough transform is used to obtain a rough estimate of the object's pose in the three-dimensional world. This pose estimate Is then refined over a linearized optimization process to achieve an optimal solution from which shape acceptability can be determined on the basis of a statistical interpretation. Experiments showed that the current inspection system is capable of registering objects having arbitrary deformations and that the system is capable of inspecting objects of complex shape.