Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
A novel feature extraction method and hybrid tree classification for handwritten numeral recognition
Autore:
Zhang, P; Chen, LH;
Indirizzi:
Nanyang Technol Univ, Sch Elect & Elect Engn, Digital Signal Proc Lab, Singapore 639798, Singapore Nanyang Technol Univ Singapore Singapore 639798 gapore 639798, Singapore
Titolo Testata:
PATTERN RECOGNITION LETTERS
fascicolo: 1-3, volume: 23, anno: 2002,
pagine: 45 - 56
SICI:
0167-8655(200201)23:1-3<45:ANFEMA>2.0.ZU;2-E
Fonte:
ISI
Lingua:
ENG
Soggetto:
CHARACTER-RECOGNITION; NEURAL-NETWORK; DIGIT RECOGNITION; CLASSIFIERS; ALGORITHM; DESIGN; INFORMATION;
Keywords:
handwritten numeral recognition; feature extraction; decision tree classifier; neural networks;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
25
Recensione:
Indirizzi per estratti:
Indirizzo: Zhang, P 2000 ST MARC,Apt 1003, Montreal, PQ H3H 2N9, Canada 2000 ST MARC,Apt 1003 Montreal PQ Canada H3H 2N9 H3H 2N9, Canada
Citazione:
P. Zhang e L.H. Chen, "A novel feature extraction method and hybrid tree classification for handwritten numeral recognition", PATT REC L, 23(1-3), 2002, pp. 45-56

Abstract

A hybrid classification system with neural network and decision tree as the classifiers for handwritten numeral recognition is proposed. Firstly a variety of stable and reliable global features are defined and extracted based on the character geometric structures, a novel floating detector is then proposed to detect segments along the left and right profiles of a character image used as local features. The recognition system consists of a hierarchical coarse classification and fine classification. For the coarse classifier: a three-layer feed forward neural network with back propagation learning algorithm is employed to distinguish six subsets {0}, {6}, {8}, {1,7}, {2, 3,5}, {4,9} based on the feature similarity of characters extracted. Three character classes namely {0}, {6} and {8} are directly recognized from artificial neural network (ANN). For each of characters in the latter threesubsets, a decision tree classifier is built for further fine classification as follows: Firstly, the specific feature-class relationship is heuristically and empirically deduced between the feature primitives and corresponding semantic class. Then, an iterative growing and pruning algorithm is used to form a tree classifier. Experiments demonstrated that the proposed recognition system is robust and flexible and a high recognition rate is reported. (C) 2002 Elsevier Science B.V. All rights reserved.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 31/03/20 alle ore 18:29:52