Metrological Characterization of a Vision-Based Measurement System for the Online Inspection of Automotive Rubber Profile

Abstract

This paper deals with the metrological characterization of a stereovision-based measurement system for the inspection of automotive rubber profiles in an industrial plant. The characterization of this class of measurement systems introduces new challenges due to both the unavailability of reference measurement instruments and the complexity of the measurement system itself, which does not allow a straightforward application of the standard procedures for uncertainty evaluation. To assign optimum values to a number of design parameters, the followed approach focuses not only on evaluating the total uncertainty but also on analyzing systematic effects and influence quantities.