OPTIMAL COMBINATIONS OF PATTERN CLASSIFIERS

被引：185

作者：

LAM, L

SUEN, CY

机构：

[1] Centre for Pattern Recognition and Machine Intelligence, Concordia University, Montreal, Que. H3G 1M8, Suite GM-606

来源：

PATTERN RECOGNITION LETTERS | 1995年 / 16卷 / 09期

关键词：

MULTIPLE CLASSIFIER SYSTEMS; OCR; MAJORITY VOTE; GENETIC ALGORITHM; BAYESIAN METHOD;

D O I：

10.1016/0167-8655(95)00050-Q

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

To improve recognition results, decisions of multiple classifiers can be combined. We study the performance of combination methods that are variations of the majority vote. A Bayesian formulation and a weighted majority vote (with weights obtained through a genetic algorithm) are implemented, and the combined performances of 7 classifiers on a large set of handwritten numerals are analyzed.

引用

页码：945 / 954

页数：10

共 16 条

[1]

FRANKE J, 1992, 11TH IAPR INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, PROCEEDINGS, VOL II, P611

[2]

GADER PD, 1990, P US POST SERV ADV T, P539

[3]

HO TK, 1994, IEEE T PATTERN ANAL, V16, P66, DOI 10.1109/34.273716

[4]

Holland J., 1989, GENETIC ALGORITHMS S

[5]

HOLLAND JH, 1975, ADAPTATION NATURAL A

[6]

HUANG YS, 1993, 3 INT WORKSH FRONT H, P11

[7]

LAM L, 1994, 12TH P INT C PATT RE, V2, P418

[8]

LAM L, 1994, 4TH P INT WORKSH FRO, P245

[9]

LEE D, 1993, 3 INT WORKSH FRONT H, P153

[10]

LIU K, 1994, P SIE C NEURAL STOCH, V3, P210

← 1 2 →