Adaptive preprocessing and segmentation for a region-based surface extraction method

Because of its great possibilities, meanwhile the industrial computer tomography (CT) is used for many different applications, where three dimensional material limitations are essential. But in many cases, the results of the conventional surface extraction method, the marching cubes algorithm, are not optimal and need to be postprocessed which is time consuming and complex. Moreover there is often no possibility to generate an adequate mesh, e.g. considering multi-material objects. In this paper we present a method for region-based surface extraction which leads to a one-step-method with optimal surface results for the next evaluation step. It is separated in a preprocessing part where the surface generation is defined by a given task or problem and in a surface extraction part where the surface is determined based on the developed preconditions. We want to overcome the disadvantages of the state-of-the-art solutions and offer a modern way of task adaptive and optimized metrology.