Detail Real Time 3d Object Recognition For Automation Robots   Oleh:  Won, Sangchul ; Jeong, Hong ; Kim, Sungjin ; Park, Sungchan Jenis: Article from Article Dalam koleksi: Final Program and Book of Abstracts: The 4th Asian Control Conference, September 25-27, 2002 (Sep. 2002), page 944-947. Topik: Automation Robots; Robot System; 3D Object Recognition Fulltext: AC021620.PDF (577.55KB) Isi artikel In this paper, we study the methodology of 3D object recognition using the disparity image, which is converted to the depth map using a simple stereo camera model. From the depth map we find the target object’s pose and position in 3-D space. In order to find the target object’s pose and position, we use the method of model-based object recognition. We compute the depth map from the model and extract the model’s and the object’s edges from a computed and real depth map. First we compute the edge errors from the model’s and object’s edges, then minimize the edge errors. When these errors are close to zero, the model’s pose and position are close to that of the object. In order to verify the accuracy of object recognition, we perform the pick and place task using a robot system Opini Anda Klik untuk menuliskan opini Anda tentang koleksi ini! Kembali

Copyright © 2006, 2007 Unika Atma Jaya, all rights reserved