Data reduction for optical 3D-inspection in automotive application

Abstract

Advances in optical sensors, computer graphics, and robotics provide new technology for the processes in styling, design, and quality management. They are used in optical inspection and reverse engineering to speed up the development process. One of the challenges of these new technologies for Daimler-Chrysler is the coupling to current CAD systems. The 3D data obtained from fast optical sensors have to be post-processed before they can be transferred to a commercial computer aided design system due to data format and the amount of the measured data. Reducing the measured data can remove important information or lower the accuracy of the data. This paper describes current and two new reduction algorithms for optical 3D data. It illustrates the need of data reduction in an application: a historic Mercedes Benz racing car is measured by a optical sensor delivering 98.3 million points. They are reduced down to 1 million points for reverse engineering of the car body surface.