Pyramid architecture classification tree

Author

Yoshii, Hiroto

Author_Institution

Canon Inc., Kawasaki, Japan

Volume

2

fYear

1996

fDate

25-29 Aug 1996

Firstpage

310

Abstract

This paper proposes a novel pattern recognition algorithm-the pyramid architecture classification tree (PACT). The learning phase of the recognition system consists of two steps: a pyramid making step and a decision tree making step; all training patterns are preprocessed by the pyramid structure and the results are used for making a decision tree. PACT directly copes with a bitmap array having the two dimensional topology and needs no feature extraction. For evaluation of the performance of PACT, various experiments using a handprint Japanese character database were carried out. The results show that PACT can realize about 50 times faster training speed than that of conventional decision tree classifiers, and classifies patterns in far higher speed than nearest neighbor matching algorithms

Keywords

character recognition; decision theory; image classification; image matching; learning (artificial intelligence); trees (mathematics); 2D topology; bitmap array; decision tree; handprint Japanese character recognition; image matching; learning phase; pattern recognition algorithm; pyramid architecture classification tree; Classification algorithms; Classification tree analysis; Decision trees; Feature extraction; Iterative algorithms; Nearest neighbor searches; Pattern recognition; Spatial databases; Testing; Topology;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition, 1996., Proceedings of the 13th International Conference on

Conference_Location

Vienna

ISSN

1051-4651

Print_ISBN

0-8186-7282-X

Type

conf

DOI

10.1109/ICPR.1996.546839

Filename

546839