Presentation 1995/9/28
Handwritten Address Recognition Algorithm with Adaptive Threshold Depending on Character Stroke Width
Hidefumi INO, Ning SUN, Yoshiaki NEMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In handwritten address recognition, it's important to segment characters correctly. The conventional methods mainly used fixed threshold to segment characters from handwritten address image. The problem of those methods is that they can not cope with variable stroke width of handwritten characters. In this paper, we propose a new algorithm in which adaptive threshold is used depending on the stroke width of character image. As a result, this algorithm obtained a high recognition rate of 89.4% for 321 test samples, distributed by Institute for Posts and Telecommunication Policy, MPT.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) recognition of handwritten address / stroke width / adaptive threshold
Paper # PRU95-109
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) Handwritten Address Recognition Algorithm with Adaptive Threshold Depending on Character Stroke Width
Sub Title (in English)
Keyword(1) recognition of handwritten address
Keyword(2) stroke width
Keyword(3) adaptive threshold
1st Author's Name Hidefumi INO
1st Author's Affiliation Graduate School of Information Sciences, Tohoku Univ.()
2nd Author's Name Ning SUN
2nd Author's Affiliation Graduate School of Information Sciences, Tohoku Univ.
3rd Author's Name Yoshiaki NEMOTO
3rd Author's Affiliation Graduate School of Information Sciences, Tohoku Univ.
Date 1995/9/28
Paper # PRU95-109
Volume (vol) vol.95
Number (no) 278
Page pp.pp.-
#Pages 6
Date of Issue