Presentation 1998/12/18
Handwritten Address Recognition Reducing Segmentation Errors
Yoshinobu Hotta, Satoshi Naoi, Misako Suwa, Junichi Hirai,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a recognitio method which utilize both key character extraction and word image recognition, as to reduce segmentation errors. First key characters which mean node of address are extracted and recognized. Then image between key characters is assumed to be one character and recognized without segmentation. In addition, based on address class knowledge, upper class address word limits the matching area of lower class word. Experimental results shows the effectivity of our method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Handwritten Address / segmentation key character / word recognition / address knowledge
Paper # PRMU98-161
Date of Issue

Conference Information
Committee PRMU
Conference Date 1998/12/18(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 Address Recognition Reducing Segmentation Errors
Sub Title (in English)
Keyword(1) Handwritten Address
Keyword(2) segmentation key character
Keyword(3) word recognition
Keyword(4) address knowledge
1st Author's Name Yoshinobu Hotta
1st Author's Affiliation FUJITSU LABORATORIES LTD.()
2nd Author's Name Satoshi Naoi
2nd Author's Affiliation FUJITSU LABORATORIES LTD.
3rd Author's Name Misako Suwa
3rd Author's Affiliation FUJITSU LABORATORIES LTD.
4th Author's Name Junichi Hirai
4th Author's Affiliation FUJITSU LTD.
Date 1998/12/18
Paper # PRMU98-161
Volume (vol) vol.98
Number (no) 490
Page pp.pp.-
#Pages 7
Date of Issue