THE PRINCIPAL COMPONENTS OF NATURAL IMAGES

被引：149

作者：

HANCOCK, PJB ^{[1
]}

BADDELEY, RJ ^{[1
]}

SMITH, LS ^{[1
]}

机构：

[1] UNIV STIRLING,CTR COGNIT & COMPUTAT NEUROSCI,DEPT COMP SCI,STIRLING FK9 4LA,SCOTLAND

来源：

NETWORK-COMPUTATION IN NEURAL SYSTEMS | 1992年 / 3卷 / 01期

关键词：

D O I：

10.1088/0954-898X/3/1/008

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A neural net was used to analyse samples of natural images and text. For the natural images, components resemble derivatives of Gaussian operators, similar to those found in visual cortex and inferred from psychophysics. While the results from natural images do not depend on scale, those from text images are highly scale dependent. Convolution of one of the text components with an original image shows that it is sensitive to inter-word gaps.

引用

页码：61 / 70

页数：10

共 15 条

[1]

BADDELEY RJ, 1991, UNPUB P R SOC B

[2]

BARROW HG, 1987, 1987 IEEE INT C NEUR, P115

[3]

BECKER S, 1989, TR CRGTR897 U TOR DE

[4] ASYMMETRIES IN ORIENTED-LINE DETECTION INDICATE 2 ORTHOGONAL FILTERS IN EARLY VISION [J].

FOSTER, DH ;

WARD, PA .

PROCEEDINGS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES, 1991, 243 (1306) :75-81

[5]

HERTZ J, 1990, SANTA FE I STUDIES S

[6] FROM BASIC NETWORK PRINCIPLES TO NEURAL ARCHITECTURE - EMERGENCE OF ORIENTATION COLUMNS [J].

LINSKER, R .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 1986, 83 (22) :8779-8783

[7] FROM BASIC NETWORK PRINCIPLES TO NEURAL ARCHITECTURE - EMERGENCE OF SPATIAL-OPPONENT CELLS [J].

LINSKER, R .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 1986, 83 (19) :7508-7512

[8] FROM BASIC NETWORK PRINCIPLES TO NEURAL ARCHITECTURE - EMERGENCE OF ORIENTATION-SELECTIVE CELLS .2. [J].

LINSKER, R .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 1986, 83 (21) :8390-8394

[9] SELF-ORGANIZATION IN A PERCEPTUAL NETWORK [J].

LINSKER, R .

COMPUTER, 1988, 21 (03) :105-117

[10] Analysis of Linsker's application of Hebbian rules to linear networks [J].

MacKay, David J. C. ;

Miller, Kenneth D. .

NETWORK-COMPUTATION IN NEURAL SYSTEMS, 1990, 1 (03) :257-297

← 1 2 →