Presentation 2007-03-15
Fundamental Study for Highly Practical OCR
Yasuji MIYAKE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) First the author's opinion about the image of the next-generation OCR is given based on the rough estimation of present commercial OCRs in view of practical usefulness and the history on the technological development of OCR. Next the weighted direction code histogram method which is one of widely used character recognition methods is described, and the background of its development, the concept, the features, and future work are discussed. At the finish, the author introduces his recent research works toward the next-generation OCR. They concern the following: the measurement of the domain covering just the characters which humans can read correctly, its application to the generation of quasi-handwritten characters and high precision recognition system of hand-written postal code, the font recognition in addition to category recognition in omni-font basis for a recognition system of formulae in publications of all ages and countries, and the detection and extraction of lines and characters from a document written in free format on white plain paper.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) OCR / Next-generation OCR / Character Recognition / Weighted Direction Code Histogram Method / Formula Recognition / Omni-font / Character Style Recognition
Paper # PRMU2006-244
Date of Issue

Conference Information
Committee PRMU
Conference Date 2007/3/8(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) Fundamental Study for Highly Practical OCR
Sub Title (in English)
Keyword(1) OCR
Keyword(2) Next-generation OCR
Keyword(3) Character Recognition
Keyword(4) Weighted Direction Code Histogram Method
Keyword(5) Formula Recognition
Keyword(6) Omni-font
Keyword(7) Character Style Recognition
1st Author's Name Yasuji MIYAKE
1st Author's Affiliation College of Engineering, Chubu University()
Date 2007-03-15
Paper # PRMU2006-244
Volume (vol) vol.106
Number (no) 605
Page pp.pp.-
#Pages 6
Date of Issue