INVARIANCE AND NEURAL NETS

被引：43

作者：

BARNARD, E ^{[1
]}

CASASENT, D ^{[1
]}

机构：

[1] CARNEGIE MELLON UNIV,DEPT QUIM ANALIT,PITTSBURGH,PA 15213

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 1991年 / 2卷 / 05期

关键词：

D O I：

10.1109/72.134287

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Invariance with respect to certain transformations is one of the main tasks of pattern-recognition systems. We study various techniques for obtaining this invariance with neural net classifiers and identify the invariant-feature technique as the most suitable for current neural classifiers. A new formulation of invariance in terms of constraints on the feature values leads to a general method for transforming any given feature space so that it becomes invariant to specified transformations. A case study using range imagery is used to exemplify these ideas, and good performance is obtained.

引用

页码：498 / 508

页数：11

共 31 条

[1] IMAGE NORMALIZATION BY COMPLEX MOMENTS [J].

ABUMOSTAFA, YS ;

PSALTIS, D .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1985, 7 (01) :46-55

[2]

Amari S., 1968, RAAG MEMOIRS, V4, P553

[3]

[Anonymous], 1987, LEARNING INTERNAL RE

[4]

Ballard DH, 1982, COMPUTER VISION

[5]

BARNARD E, 1990, APPL OPTICS, V29, P2603

[6]

BARNARD E, 1989, IEEE T SYST MAN CYBE, V19

[7] A NEURAL NETWORK FOR INVARIANT PATTERN-RECOGNITION [J].

BIENENSTOCK, E ;

VONDERMALSBURG, C .

EUROPHYSICS LETTERS, 1987, 4 (01) :121-126

[8] EXPERIMENTS ON NEURAL NET RECOGNITION OF SPOKEN AND WRITTEN TEXT [J].

BURR, DJ .

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1988, 36 (07) :1162-1168

[9] COMPUTER GENERATED HOLOGRAMS IN PATTERN-RECOGNITION - A REVIEW [J].

CASASENT, D .

OPTICAL ENGINEERING, 1985, 24 (05) :724-730

[10] UNIFIED SYNTHETIC DISCRIMINANT FUNCTION COMPUTATIONAL FORMULATION [J].

CASASENT, D .

APPLIED OPTICS, 1984, 23 (10) :1620-1627

← 1 2 3 4 →