Maximum mutual information estimation of a simplified hidden MRF for off-line hand-written Chinese characters recognition

Research output: Contribution to journalConference article

Abstract

Understanding of hand-written Chinese characters is at such a primitive stage that models include some assumptions about hand-written Chinese characters that are simply false. So Maximum Likelihood Estimation (MLE) may not be an optimal method for hand-written Chinese characters recognition. This concern motivates the research effort to consider alternative criteria. Maximum Mutual Information Estimation (MMIE) is an alternative method for parameter estimation that does not derive its rationale from presumed model correctness, but instead examines the pattern-modeling problem in automatic recognition system from an information-theoretic point of view. The objective of MMIE is to find a set of parameters in such that the resultant model allows the system to derive from the observed data as much information as possible about the class. We consider MMIE for recognition of hand-written Chinese characters using on a simplified hidden Markov Random Field (MRF). MMIE provides improved performance improvement over MLE in this application.

Original languageEnglish (US)
Pages (from-to)58-63
Number of pages6
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume3651
StatePublished - Jan 1 1999
EventProceedings of the 1999 6th Annual Conference on Document Recognition and Retrieval VI - San Jose, CA, USA
Duration: Jan 27 1999Jan 28 1999

Fingerprint

character recognition
Character recognition
Character Recognition
Mutual Information
Random Field
Line
Maximum likelihood estimation
Maximum Likelihood Estimation
Alternatives
Parameter estimation
Parameter Estimation
Correctness
Model
Modeling
Character

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

@article{fe33bb44d1c94a0d9c3a55f98111afa9,
title = "Maximum mutual information estimation of a simplified hidden MRF for off-line hand-written Chinese characters recognition",
abstract = "Understanding of hand-written Chinese characters is at such a primitive stage that models include some assumptions about hand-written Chinese characters that are simply false. So Maximum Likelihood Estimation (MLE) may not be an optimal method for hand-written Chinese characters recognition. This concern motivates the research effort to consider alternative criteria. Maximum Mutual Information Estimation (MMIE) is an alternative method for parameter estimation that does not derive its rationale from presumed model correctness, but instead examines the pattern-modeling problem in automatic recognition system from an information-theoretic point of view. The objective of MMIE is to find a set of parameters in such that the resultant model allows the system to derive from the observed data as much information as possible about the class. We consider MMIE for recognition of hand-written Chinese characters using on a simplified hidden Markov Random Field (MRF). MMIE provides improved performance improvement over MLE in this application.",
author = "Yah Xiong and Reichenbach, {Stephen E.}",
year = "1999",
month = "1",
day = "1",
language = "English (US)",
volume = "3651",
pages = "58--63",
journal = "Proceedings of SPIE - The International Society for Optical Engineering",
issn = "0277-786X",
publisher = "SPIE",

}

TY - JOUR

T1 - Maximum mutual information estimation of a simplified hidden MRF for off-line hand-written Chinese characters recognition

AU - Xiong, Yah

AU - Reichenbach, Stephen E.

PY - 1999/1/1

Y1 - 1999/1/1

N2 - Understanding of hand-written Chinese characters is at such a primitive stage that models include some assumptions about hand-written Chinese characters that are simply false. So Maximum Likelihood Estimation (MLE) may not be an optimal method for hand-written Chinese characters recognition. This concern motivates the research effort to consider alternative criteria. Maximum Mutual Information Estimation (MMIE) is an alternative method for parameter estimation that does not derive its rationale from presumed model correctness, but instead examines the pattern-modeling problem in automatic recognition system from an information-theoretic point of view. The objective of MMIE is to find a set of parameters in such that the resultant model allows the system to derive from the observed data as much information as possible about the class. We consider MMIE for recognition of hand-written Chinese characters using on a simplified hidden Markov Random Field (MRF). MMIE provides improved performance improvement over MLE in this application.

AB - Understanding of hand-written Chinese characters is at such a primitive stage that models include some assumptions about hand-written Chinese characters that are simply false. So Maximum Likelihood Estimation (MLE) may not be an optimal method for hand-written Chinese characters recognition. This concern motivates the research effort to consider alternative criteria. Maximum Mutual Information Estimation (MMIE) is an alternative method for parameter estimation that does not derive its rationale from presumed model correctness, but instead examines the pattern-modeling problem in automatic recognition system from an information-theoretic point of view. The objective of MMIE is to find a set of parameters in such that the resultant model allows the system to derive from the observed data as much information as possible about the class. We consider MMIE for recognition of hand-written Chinese characters using on a simplified hidden Markov Random Field (MRF). MMIE provides improved performance improvement over MLE in this application.

UR - http://www.scopus.com/inward/record.url?scp=0032675414&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032675414&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:0032675414

VL - 3651

SP - 58

EP - 63

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

ER -