Options
2004
Conference Paper
Titel
Automated segmentation and object recognition in 3D data sets
Abstract
Best fitting of geometric primitives is in many fields of science and engineering, e.g. metrology, reverse engineering or computer vision, an important issue. Nowadays spatial data sets are generated using laser-systems, computer-tomography, coordinate measuring devices (CMM) or stereovision, depending on the application and desired accuracy. Due to the progress in sensor technology large data sets will be available in real time with low-cost sensors for a wide range of applications, so that the acquisition of spatial data sets is accelerated and becomes less expensive. As a consequence the various areas of applications for segmentation and object recognition methods will be further enlarged and powerful algorithms to cope with large data sets will be needed. In this paper we present a very robust and run-time efficient method for automated object recognition. The 3D data sets, which can be processed by the algorithm, do not have to fulfil special requirements. Moreover the segmentation is proceeded automatically and goes hand in hand with feature extraction. Accordingly there is no neccessity for user interaction any more. The algorithms are applied to real point clouds and the results are shown. As a conclusion future work and perspectives of the technology will be discussed.