王建刚, 王寻羽, 白雪生, 徐心平. 基于神经网络的三维物体姿态测定[J]. 机器人, 1996, 18(2): 83-90..
WANG Jiangang, WANG Xunyu, BAI Xuesheng, XU Xinping. APPLICATION OF HOPFIELD NEURAL NETWORK TO DETERMINE THE POSE OF 3-D OBJECTS. ROBOT, 1996, 18(2): 83-90..
Abstract:A method for location of 3-D objects is introduced. 3-D objects attitude is first determined by the interpretation of three image lines as the perspective projection of three linear ridges of the object model. Then a Hopfield neural network is applied for determinig the pose of 3-D objects.
[1] Robert Bergevin, Martin D Levine. Generic Object Recognition,Building and Matching Coarse Descriptions from Line Drawings. IEEE Trans on PAMI, 1993,15(1):19-36
[2] Brooks R A. Symbolic Reasoning Among 3-D Models and 2-D Images. Artif Intell. 1981,1-7, 285348
[3] 王建刚,姚筱亦.三维物体识别及姿态侧定的推理系统.机器人,1994.16(2), 104 113
[4] Wang liangang. Yao Xiaoyi. An Intelligent Method for Inference and Extraction of 3-D Information of Objects. Proc 2nd Asia Conf on Robotics and Its Application, Beijing, China,1994,660-664
[5] Ganapathy S. Decomposition of Transformation Matrices for Robot Vision. Proc IEEE Conf on Robotics. Atlanta,CA, 1984:130-139
[6] Michel D, Marc R. Lean-Thierry L. Gerard R. Determination o1 the Attitude of 3-D Objects from a Single Perspective View. IEEE Trana on PAMf. 1989,11(12):1265-1278
[7] 姚筱亦.处理单目立体视觉问题的新构思.第一届全国智能信息技术讨论会,北京.1992
[8] Kumar S, Ray D, Dutta Majumder. Application of Hopfield Neural Networks and Canonical Perspectives to Recognize and Locate Partially Occluded 3-D Ob沁ts. Pattern Recognition Letters, 1994,15, 815-824