AN ULTRASONIC VISUAL SENSOR FOR 3-DIMENSIONAL OBJECT RECOGNITION USING NEURAL NETWORKS

被引:39
作者
WATANABE, S
YONEYAMA, M
机构
[1] Research and Development Center, Ricoh Co. Ltd., Kohoku-ku, Yokohama, 223, 16–1, Shin-ei-cho
来源
IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION | 1992年 / 8卷 / 02期
关键词
D O I
10.1109/70.134277
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
By combining ultrasonic imaging with neural networks, we have developed a new 3-D object recognition system for use in robotic vision. Ultrasonic imaging is used to calculate the initial 3-D images of the objects. These images are then passed to neural networks that 1) identify object categories, 2) estimate object locations, and 3) improve 3-D images. In this paper, we explain the 3-D ultrasonic imaging method, propose three neural network structures for 3-D image analysis, and demonstrate the practicability of the system through experimental results, which show that quick and accurate recognition can be achieved using only a small set of transducers.
引用
收藏
页码:240 / 249
页数:10
相关论文
共 48 条
[1]   3-DIMENSIONAL ULTRASONIC VISION FOR ROBOTIC APPLICATIONS [J].
ACAMPORA, AS ;
WINTERS, JH .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1989, 11 (03) :291-303
[2]  
BREEUWER R, 1989, P IEEE ULTR S, P635
[3]  
Brown M.K., 1985, IEEE T ROBOTIC AUTOM, P110
[4]   ULTRASONIC SURFACE IMAGING IN ADVERSE ENVIRONMENTS [J].
BRYANT, RC ;
BOGNER, RE .
IEEE TRANSACTIONS ON SONICS AND ULTRASONICS, 1984, 31 (04) :373-390
[5]  
Chen Y.-C., 1987, Proceedings of IECON '87: 1987 International Conference on Industrial Electronics, Control, and Instrumentation (Cat. No.CH2484-4), P690
[6]  
CONRATH BC, 1989, P IEEE ULTRASON S, P1007
[7]  
COTTREL GW, 1987, IMAGE COMPRESSION BA
[8]  
Cybenko G., 1989, Mathematics of Control, Signals, and Systems, V2, P303, DOI 10.1007/BF02551274
[9]  
DORF RC, 1984, 8TH P ROB C, V2, P44
[10]  
Duda R. O., 1973, PATTERN CLASSIFICATI, V3