A bottom-up and top-down approach to using context in text recognition

Radian Shinghal, Godfried Toussaint

Research output: Contribution to journalArticle

Abstract

Existing approaches to using contextual information in text recognition tend to fall into two categories: dictionary look-up methods and Markov methods. Markov methods use transition probabilities between letters and represent a bottom-up approach to using context which is characterized by being very efficient but exhibiting mediocre errorcorrecting capability. Dictionary look-up methods, on the other hand, constrain the choice of letter sequences to be legal words and represent a top-down approach characterized by impressive error-correcting capabilities at a stiff price in storage and computation. In this paper, a combined bottom-up top-down algorithm is proposed. Exhaustive experimentation shows that the algorithm achieves the error-correcting capability of the dictionary look-up methods at half the cost.

Original languageEnglish (US)
Pages (from-to)201-212
Number of pages12
JournalInternational Journal of Man-Machine Studies
Volume11
Issue number2
DOIs
StatePublished - Jan 1 1979

Fingerprint

Glossaries
dictionary
Costs
costs

ASJC Scopus subject areas

  • Engineering(all)

Cite this

A bottom-up and top-down approach to using context in text recognition. / Shinghal, Radian; Toussaint, Godfried.

In: International Journal of Man-Machine Studies, Vol. 11, No. 2, 01.01.1979, p. 201-212.

Research output: Contribution to journalArticle

@article{17acec163c9f42789171491299c919ae,
title = "A bottom-up and top-down approach to using context in text recognition",
abstract = "Existing approaches to using contextual information in text recognition tend to fall into two categories: dictionary look-up methods and Markov methods. Markov methods use transition probabilities between letters and represent a bottom-up approach to using context which is characterized by being very efficient but exhibiting mediocre errorcorrecting capability. Dictionary look-up methods, on the other hand, constrain the choice of letter sequences to be legal words and represent a top-down approach characterized by impressive error-correcting capabilities at a stiff price in storage and computation. In this paper, a combined bottom-up top-down algorithm is proposed. Exhaustive experimentation shows that the algorithm achieves the error-correcting capability of the dictionary look-up methods at half the cost.",
author = "Radian Shinghal and Godfried Toussaint",
year = "1979",
month = "1",
day = "1",
doi = "10.1016/S0020-7373(79)80017-6",
language = "English (US)",
volume = "11",
pages = "201--212",
journal = "International Journal of Human Computer Studies",
issn = "1071-5819",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - A bottom-up and top-down approach to using context in text recognition

AU - Shinghal, Radian

AU - Toussaint, Godfried

PY - 1979/1/1

Y1 - 1979/1/1

N2 - Existing approaches to using contextual information in text recognition tend to fall into two categories: dictionary look-up methods and Markov methods. Markov methods use transition probabilities between letters and represent a bottom-up approach to using context which is characterized by being very efficient but exhibiting mediocre errorcorrecting capability. Dictionary look-up methods, on the other hand, constrain the choice of letter sequences to be legal words and represent a top-down approach characterized by impressive error-correcting capabilities at a stiff price in storage and computation. In this paper, a combined bottom-up top-down algorithm is proposed. Exhaustive experimentation shows that the algorithm achieves the error-correcting capability of the dictionary look-up methods at half the cost.

AB - Existing approaches to using contextual information in text recognition tend to fall into two categories: dictionary look-up methods and Markov methods. Markov methods use transition probabilities between letters and represent a bottom-up approach to using context which is characterized by being very efficient but exhibiting mediocre errorcorrecting capability. Dictionary look-up methods, on the other hand, constrain the choice of letter sequences to be legal words and represent a top-down approach characterized by impressive error-correcting capabilities at a stiff price in storage and computation. In this paper, a combined bottom-up top-down algorithm is proposed. Exhaustive experimentation shows that the algorithm achieves the error-correcting capability of the dictionary look-up methods at half the cost.

UR - http://www.scopus.com/inward/record.url?scp=0018449051&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0018449051&partnerID=8YFLogxK

U2 - 10.1016/S0020-7373(79)80017-6

DO - 10.1016/S0020-7373(79)80017-6

M3 - Article

AN - SCOPUS:0018449051

VL - 11

SP - 201

EP - 212

JO - International Journal of Human Computer Studies

JF - International Journal of Human Computer Studies

SN - 1071-5819

IS - 2

ER -