Adaptive confidence transform based classifier combination for Chinese character recognition

被引：60

作者：

Lin, XF ^{[1
]}

Ding, XQ ^{[1
]}

Chen, M ^{[1
]}

Zhang, R ^{[1
]}

Wu, YS ^{[1
]}

机构：

[1] Tsinghua Univ, Dept Elect Engn, Image Proc Div, Beijing 100084, Peoples R China

来源：

PATTERN RECOGNITION LETTERS | 1998年 / 19卷 / 10期

基金：

美国国家科学基金会;

关键词：

printed Chinese character recognition; handwritten Chinese character recognition; multiple classifier combination; consensus theory; a posteriori probability;

D O I：

10.1016/S0167-8655(98)00072-5

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Classifier combination is an effective way to improve recognition performance. However, in Chinese character recognition the extremely large number of categories results in several difficulties for the combination. In order to overcome these difficulties a novel combination method is presented in this paper. It consists of three main components: adaptive confidence transform (ACT), consensus theoretic combination and reliability based speedup scheme. ACT, which can estimate a posteriori probabilities from raw measurement values, is the focus of this paper. Experimental results show a significant reduction of error rates in both printed (PCCR) and handwritten Chinese character recognition (HCCR). (C) 1998 Elsevier Science B.V. All rights reserved.

引用

页码：975 / 988

页数：14

共 18 条

[1] RECOGNITION TECHNOLOGY FRONTIERS [J].

BAIRD, HS .

PATTERN RECOGNITION LETTERS, 1993, 14 (04) :327-334

[2] DEMOCRACY IN NEURAL NETS - VOTING SCHEMES FOR CLASSIFICATION [J].

BATTITI, R ;

COLLA, AM .

NEURAL NETWORKS, 1994, 7 (04) :691-707

[3]

BENDEDIKTSSON JA, 1992, IEEE T SYST MAN CYB, V22, P688

[4]

CHEN Y, 1997, THESIS TSINGHUA U CH

[5]

Chen YB, 1997, PROC INT CONF DOC, P206, DOI 10.1109/ICDAR.1997.619842

[6]

GUO H, 1949, THESIS TSINGHUA U CH

[7]

GUO H, 1996, P 1 C MULT INT BEIJ, P221

[8] OPTICAL RECOGNITION OF HANDWRITTEN CHINESE CHARACTERS - ADVANCES SINCE 1980 [J].

HILDEBRANDT, TH ;

LIU, WT .

PATTERN RECOGNITION, 1993, 26 (02) :205-225

[9]

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

[10]

Huang Y. S., 1993, Proceedings of the Second International Conference on Document Analysis and Recognition (Cat. No.93TH0578-5), P598, DOI 10.1109/ICDAR.1993.395664

← 1 2 →