Automatic classification of spatial relationships among mathematical symbols using geometric features

Walaa Aly, Seiichi Uchida, Masakazu Suzuki

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Machine recognition of mathematical expressions on printed documents is not trivial even when all the individual characters and symbols in an expression can be recognized correctly. In this paper, an automatic classification method of spatial relationships between the adjacent symbols in a pair is presented. This classification is important to realize an accurate structure analysis module of math OCR. Experimental results on very large databases showed that this classification worked well with an accuracy of 99.525% by using distribution maps which are defined by two geometric features, relative size and relative position, with careful treatment on document-dependent characteristics.

Original languageEnglish
Pages (from-to)2235-2243
Number of pages9
JournalIEICE Transactions on Information and Systems
VolumeE92-D
Issue number11
DOIs
Publication statusPublished - Jan 1 2009

Fingerprint

Optical character recognition

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

Automatic classification of spatial relationships among mathematical symbols using geometric features. / Aly, Walaa; Uchida, Seiichi; Suzuki, Masakazu.

In: IEICE Transactions on Information and Systems, Vol. E92-D, No. 11, 01.01.2009, p. 2235-2243.

Research output: Contribution to journalArticle

@article{2d49934504794f24beb97feb746754c8,
title = "Automatic classification of spatial relationships among mathematical symbols using geometric features",
abstract = "Machine recognition of mathematical expressions on printed documents is not trivial even when all the individual characters and symbols in an expression can be recognized correctly. In this paper, an automatic classification method of spatial relationships between the adjacent symbols in a pair is presented. This classification is important to realize an accurate structure analysis module of math OCR. Experimental results on very large databases showed that this classification worked well with an accuracy of 99.525{\%} by using distribution maps which are defined by two geometric features, relative size and relative position, with careful treatment on document-dependent characteristics.",
author = "Walaa Aly and Seiichi Uchida and Masakazu Suzuki",
year = "2009",
month = "1",
day = "1",
doi = "10.1587/transinf.E92.D.2235",
language = "English",
volume = "E92-D",
pages = "2235--2243",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "一般社団法人電子情報通信学会",
number = "11",

}

TY - JOUR

T1 - Automatic classification of spatial relationships among mathematical symbols using geometric features

AU - Aly, Walaa

AU - Uchida, Seiichi

AU - Suzuki, Masakazu

PY - 2009/1/1

Y1 - 2009/1/1

N2 - Machine recognition of mathematical expressions on printed documents is not trivial even when all the individual characters and symbols in an expression can be recognized correctly. In this paper, an automatic classification method of spatial relationships between the adjacent symbols in a pair is presented. This classification is important to realize an accurate structure analysis module of math OCR. Experimental results on very large databases showed that this classification worked well with an accuracy of 99.525% by using distribution maps which are defined by two geometric features, relative size and relative position, with careful treatment on document-dependent characteristics.

AB - Machine recognition of mathematical expressions on printed documents is not trivial even when all the individual characters and symbols in an expression can be recognized correctly. In this paper, an automatic classification method of spatial relationships between the adjacent symbols in a pair is presented. This classification is important to realize an accurate structure analysis module of math OCR. Experimental results on very large databases showed that this classification worked well with an accuracy of 99.525% by using distribution maps which are defined by two geometric features, relative size and relative position, with careful treatment on document-dependent characteristics.

UR - http://www.scopus.com/inward/record.url?scp=77950259338&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950259338&partnerID=8YFLogxK

U2 - 10.1587/transinf.E92.D.2235

DO - 10.1587/transinf.E92.D.2235

M3 - Article

AN - SCOPUS:77950259338

VL - E92-D

SP - 2235

EP - 2243

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 11

ER -