Real-time classification of industrial products based on the photonic-mixer-device sensor technology

Abstract

During the last years, Time-of-Flight sensors achieved a significant impact onto research fields in machine vision. In comparison to stereo vision system and laser range scanners they combine the advantages of active sensors providing accurate distance measurements and camera-based systems recording a 2D matrix at a high frame rate. Moreover low cost 3D imaging has the potential to open a wide field of additional applications and solutions in markets like consumer electronics, multimedia, digital photography, robotics and medical technologies. This paper focuses on the classification task of typical industrial products in the close-up range. A new approach due to the use of the photonic-mixer-device (PMD) sensor technology is presented which achieves a much simpler real-time classification. Experimental results show that the proposed approach is well suited for machine vision applications.