Title :
World-centered representation for neural networks
Author_Institution :
Dept. of Autom. Control, Northwestern Polytech. Univ., Xian, China
Abstract :
A basic problem for pattern recognition is invariance. This article argues that the traditional full-parallel computation of neural networks that has excluded many useful series computation methods seems improper for the invariance. We introduce a series search mechanism into neural computing. The world-centered recognition is acquired by a pattern-centered memory plus a space search
Keywords :
feature extraction; invariance; neural nets; object recognition; pattern recognition; search problems; invariance; neural networks; pattern recognition; pattern-centered memory; series search mechanism; space search; world-centered representation; Artificial neural networks; Automatic control; Biological system modeling; Biology computing; Computer networks; Concurrent computing; Fourier transforms; Neural networks; Pattern matching; Pattern recognition;
Conference_Titel :
Neural Networks,1997., International Conference on
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.611737
