SIMULTANEOUS LEARNING OF DECISION RULES AND IMPORTANT ATTRIBUTES FOR CLASSIFICATION PROBLEMS IN IMAGE-ANALYSIS

被引：11

作者：

PUDIL, P ^{[1
]}

NOVOVICOVA, J ^{[1
]}

KITTLER, J ^{[1
]}

机构：

[1] CZECHOSLOVAK ACAD SCI,INST INFORMAT THEORY & AUTOMAT,CS-18208 PRAGUE 8,CZECH REPUBLIC

来源：

IMAGE AND VISION COMPUTING | 1994年 / 12卷 / 03期

关键词：

FEATURE SELECTION; CLASSIFICATION; PDF ESTIMATION;

D O I：

10.1016/0262-8856(94)90072-8

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A new method for automatic machine learning of decision rules for classification problems in image analysis is presented. The method facilitates simultaneous decision rule inference and selection of discriminative features which characterize the image entities to be classified. The basis of the method is to approximate class conditional densities by a mixture of parameterized densities. Its performance is tested on a classification problem involving real image data.

引用

页码：193 / 198

页数：6

共 14 条

[1] MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].

DEMPSTER, AP ;

LAIRD, NM ;

RUBIN, DB .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38

[2]

Devijver PA, 1982, PATTERN RECOGNITION

[3]

FUKUNAGA K, 1990, INTRO STATISTICAL PA

[4]

GRIM J, 1986, KYBERNETIKA, V22, P142

[5]

HAND DJ, 1981, DISCRIMINATION CLASS, pCH6

[6] PATTERNS IN PATTERN-RECOGNITION - 1968-1974 [J].

KANAL, L .

IEEE TRANSACTIONS ON INFORMATION THEORY, 1974, 20 (06) :697-722

[7] APPROXIMATING DISCRETE PROBABILITY DISTRIBUTIONS [J].

KU, HH ;

KULLBACK, S .

IEEE TRANSACTIONS ON INFORMATION THEORY, 1969, 15 (04) :444-+

[8] ON THE USE OF I-DIVERGENCE FOR GENERATING DISTRIBUTION APPROXIMATIONS [J].

MAIA, MAGM ;

FAIRHURST, MC .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1983, 5 (06) :661-664

[9]

PUDIL P, 1991, KYBERNETIKA S, V27, P1

[10]

PUDIL P, UNPUB PATT RECOGN

← 1 2 →