Inspection of specular and painted surfaces with centralized fusion techniques

Abstract

This contribution presents new strategies to inspect specular and painted surfaces. Structures on such surfaces are normally only visible, if patterns of the environment are reflected in them. Thus, conventional approaches only yield a little information gain from a single measurement. In our approach, different intensity patterns are systematically generated in the environment of the surface such that these are reflected in the surface and captured by a camera. Following, the recorded images are processed simultaneously by a centralized fusion technique. Since the fused information is closer to the source, a better exploitation of the raw data is achieved. The fusion problem is formulated with an energy function. Its minimization yields the desired surface defects. The methodology is illustrated with two case studies: the analysis of machined surfaces, and the inspection of painted free-form surfaces. In both cases, a reliable yet cost-efficient inspection is attained matching the needs of industry.