Presentation 2014-12-19
Improving OCR accuracy of reading printed characters on the noisy background
Shogo YASUDA, Tomoaki KIMURA, Hiroyuki TSUJI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) When reading characters printed on the background of fine patterns with OCR, the binarization process often fails and generates spike-like and/or line-shaped noise in the binary results, which makes the OCR reading accuracy significantly reduced. In this report, we proposed a pre-processing scheme that can effectively remove spike-like noise without altering the silhouette of each character in order to improve the OCR reading accuracy. The effectiveness of the proposed method was confirmed by applying the method to an open source OCR engine, named "Tesseract".
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Optical character recognition / OCR / Serial number recognition / Impulse noise removal
Paper # SIS2014-82
Date of Issue

Conference Information
Committee SIS
Conference Date 2014/12/11(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 Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Improving OCR accuracy of reading printed characters on the noisy background
Sub Title (in English)
Keyword(1) Optical character recognition
Keyword(2) OCR
Keyword(3) Serial number recognition
Keyword(4) Impulse noise removal
1st Author's Name Shogo YASUDA
1st Author's Affiliation Kanagawa Institute of Technology()
2nd Author's Name Tomoaki KIMURA
2nd Author's Affiliation Kanagawa Institute of Technology
3rd Author's Name Hiroyuki TSUJI
3rd Author's Affiliation Kanagawa Institute of Technology
Date 2014-12-19
Paper # SIS2014-82
Volume (vol) vol.114
Number (no) 370
Page pp.pp.-
#Pages 5
Date of Issue