A connectionist expert system model for conflict resolution in unconstrained handwritten numeral recognition

N. V. Subba Reddy, P. Nagabhushan

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The paper describes a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions identified by the neural network. The neural network classifier is a combination of Modified Self-Organizing Map (MSOM) and Learning Vector Quantization (LVQ). The experiments are conducted on Self-Organizing Map (SOM), MSOM, and the combinations of SOM & LVQ and MSOM & LVQ techniques. These experiments on a database of unconstrained handwritten samples show that the combination of MSOM & LVQ achieves satisfactory results in terms of classification, recognition and training time. The results obtained from this two-tier architecture are compared with the comments collected from an experiment conducted with a group of human experts specialized in unconstrained handwritten character recognition. The developed system is found to be useful in resolving conflicts in the recognition of Unconstrained Handwritten Numerals of PIN or ZIP codes of mailing addresses.

Original languageEnglish
Pages (from-to)161-169
Number of pages9
JournalPattern Recognition Letters
Volume19
Issue number2
Publication statusPublished - 01-02-1998

Fingerprint

Self organizing maps
Expert systems
Vector quantization
Neural networks
Character recognition
Experiments
Classifiers

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

@article{bf440387a3ba4f8ba7096a7843fc716c,
title = "A connectionist expert system model for conflict resolution in unconstrained handwritten numeral recognition",
abstract = "The paper describes a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions identified by the neural network. The neural network classifier is a combination of Modified Self-Organizing Map (MSOM) and Learning Vector Quantization (LVQ). The experiments are conducted on Self-Organizing Map (SOM), MSOM, and the combinations of SOM & LVQ and MSOM & LVQ techniques. These experiments on a database of unconstrained handwritten samples show that the combination of MSOM & LVQ achieves satisfactory results in terms of classification, recognition and training time. The results obtained from this two-tier architecture are compared with the comments collected from an experiment conducted with a group of human experts specialized in unconstrained handwritten character recognition. The developed system is found to be useful in resolving conflicts in the recognition of Unconstrained Handwritten Numerals of PIN or ZIP codes of mailing addresses.",
author = "{Subba Reddy}, {N. V.} and P. Nagabhushan",
year = "1998",
month = "2",
day = "1",
language = "English",
volume = "19",
pages = "161--169",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",
number = "2",

}

A connectionist expert system model for conflict resolution in unconstrained handwritten numeral recognition. / Subba Reddy, N. V.; Nagabhushan, P.

In: Pattern Recognition Letters, Vol. 19, No. 2, 01.02.1998, p. 161-169.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A connectionist expert system model for conflict resolution in unconstrained handwritten numeral recognition

AU - Subba Reddy, N. V.

AU - Nagabhushan, P.

PY - 1998/2/1

Y1 - 1998/2/1

N2 - The paper describes a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions identified by the neural network. The neural network classifier is a combination of Modified Self-Organizing Map (MSOM) and Learning Vector Quantization (LVQ). The experiments are conducted on Self-Organizing Map (SOM), MSOM, and the combinations of SOM & LVQ and MSOM & LVQ techniques. These experiments on a database of unconstrained handwritten samples show that the combination of MSOM & LVQ achieves satisfactory results in terms of classification, recognition and training time. The results obtained from this two-tier architecture are compared with the comments collected from an experiment conducted with a group of human experts specialized in unconstrained handwritten character recognition. The developed system is found to be useful in resolving conflicts in the recognition of Unconstrained Handwritten Numerals of PIN or ZIP codes of mailing addresses.

AB - The paper describes a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions identified by the neural network. The neural network classifier is a combination of Modified Self-Organizing Map (MSOM) and Learning Vector Quantization (LVQ). The experiments are conducted on Self-Organizing Map (SOM), MSOM, and the combinations of SOM & LVQ and MSOM & LVQ techniques. These experiments on a database of unconstrained handwritten samples show that the combination of MSOM & LVQ achieves satisfactory results in terms of classification, recognition and training time. The results obtained from this two-tier architecture are compared with the comments collected from an experiment conducted with a group of human experts specialized in unconstrained handwritten character recognition. The developed system is found to be useful in resolving conflicts in the recognition of Unconstrained Handwritten Numerals of PIN or ZIP codes of mailing addresses.

UR - http://www.scopus.com/inward/record.url?scp=0031995319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031995319&partnerID=8YFLogxK

M3 - Article

VL - 19

SP - 161

EP - 169

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 2

ER -