Presentation 1995/9/28
Segmentation and Recognition of Handwritten Characters by Multiple Character Subspaces
Yuji Motegi, Toshihiro Teranishi, Yasuo Ariki,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Segmentation of characters freely written on the paper is a difficult problem for a computer system. We propose in this paper the method to solve this problem by computing likelihood of characters in segmentation process based on a subspace method. At first, a binary image on which characters are written, is scanned by a fixed size of a window. At every scanning location, feature vectors are obtained and projected to each character subspace. The character recognition using subspace method is carried out and character name and its confidence are obtained. The confidence is computed as the length of the normalized projection vector to the character subspace. In this paper, character segmentation and recognition experiments were carried out using mesh feature as well as direction histogram for isolatedly, normally and connectedly written characters. As a result, one subspace of mesh feature with one dimension showed high accuracy of character segmentation for isolatedly and normally written characters. For connectedly written characters, multiple subspaces of direction histogram improved character segmentation rate.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Handwritten character / Character recognition / Character segmentation / Subspace method
Paper # PRU95-105
Date of Issue

Conference Information
Committee PRU
Conference Date 1995/9/28(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 Understanding (PRU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Segmentation and Recognition of Handwritten Characters by Multiple Character Subspaces
Sub Title (in English)
Keyword(1) Handwritten character
Keyword(2) Character recognition
Keyword(3) Character segmentation
Keyword(4) Subspace method
1st Author's Name Yuji Motegi
1st Author's Affiliation Faculty of Science and Techno1ogy, Ryukoku University()
2nd Author's Name Toshihiro Teranishi
2nd Author's Affiliation Faculty of Science and Techno1ogy, Ryukoku University
3rd Author's Name Yasuo Ariki
3rd Author's Affiliation Faculty of Science and Techno1ogy, Ryukoku University
Date 1995/9/28
Paper # PRU95-105
Volume (vol) vol.95
Number (no) 278
Page pp.pp.-
#Pages 6
Date of Issue