Presentation 1997/3/19
Handwritten Numeral Recognition using Gradient and Curvature of Gray Scale Image
Tetsushi WAKABAYASHI, Shinji TSURUOKA, Fumitaka KIMURA, Yasuji MIYAKE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this report,accuracy improvement of handwritten numeral recognition, by adding the curvature feature obtained from gray scale contour to the direction feature of the gradient, is discussed.For dimension reduction of feature vector, the FKL(Fisher-Karhunen-Loeve)method, a generalization of the canonical discriminant analysis and the principal component analysis is shown to be most effective. The high recognition rate of 99.30% is achieved for handwritten numeral data base "IPT CDROM1" distributed by IPTP(Institute for Posts and Telecommunications Policy).
Keyword(in Japanese) (See Japanese page)
Keyword(in English) gradient / curvature / ZIP code / numeral recognition / feature extraction / FKL
Paper # PRMU96-204
Date of Issue

Conference Information
Committee PRMU
Conference Date 1997/3/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Handwritten Numeral Recognition using Gradient and Curvature of Gray Scale Image
Sub Title (in English)
Keyword(1) gradient
Keyword(2) curvature
Keyword(3) ZIP code
Keyword(4) numeral recognition
Keyword(5) feature extraction
Keyword(6) FKL
1st Author's Name Tetsushi WAKABAYASHI
1st Author's Affiliation Department of Information Engineering()
2nd Author's Name Shinji TSURUOKA
2nd Author's Affiliation Department of Electrical and Electronic Engineering, Faculty of Engineering, Mie University
3rd Author's Name Fumitaka KIMURA
3rd Author's Affiliation Department of Information Engineering
4th Author's Name Yasuji MIYAKE
4th Author's Affiliation Department of Information Engineering
Date 1997/3/19
Paper # PRMU96-204
Volume (vol) vol.96
Number (no) 599
Page pp.pp.-
#Pages 8
Date of Issue