Off-line recognition of totally unconstrained handwritten numerals using multilayer cluster neural network

Research output: Contribution to journalArticle

116 Citations (Scopus)

Abstract

In this paper, we propose a new scheme for off-line recognition of totally unconstrained handwritten numerals using a simple multilayer cluster neural network trained with the back propagation algorithm and show that the use of genetic algorithms avoids the problem of finding local minima in training the multilayer cluster neural network with gradient descent technique, and improves the recognition rates. In the proposed scheme, Kirsch masks are adopted for extracting feature vectors and a three-layer cluster neural network with five independent subnetworks is developed for classifying similar numerals efficiently. In order to verify the performance of the proposed multilayer cluster neural network, experiments with handwritten numeral database of Concordia University of Canada, that of Electro-Technical Laboratory of Japan, and that of Electronics and Telecommunications Research Institute of Korea were performed. For the case of determining the initial weights using a genetic algorithm, 97.10%, 99.12%, and 99.40% correct recognition rates were obtained, respectively.

Original languageEnglish
Pages (from-to)648-652
Number of pages5
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume18
Issue number6
DOIs
Publication statusPublished - 1996 Dec 1

Fingerprint

Numeral
Multilayer
Multilayers
Neural Networks
Neural networks
Line
Telecommunications
Genetic algorithms
Genetic Algorithm
Korea
Masks
Canada
Backpropagation algorithms
Back-propagation Algorithm
Gradient Descent
Japan
Feature Vector
Local Minima
Databases
Mask

Keywords

  • Genetic algorithm
  • Multilayer cluster neural network
  • Totally unconstrained handwritten numeral recognition

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

@article{b3a291b9d19847a2880f80772e3df807,
title = "Off-line recognition of totally unconstrained handwritten numerals using multilayer cluster neural network",
abstract = "In this paper, we propose a new scheme for off-line recognition of totally unconstrained handwritten numerals using a simple multilayer cluster neural network trained with the back propagation algorithm and show that the use of genetic algorithms avoids the problem of finding local minima in training the multilayer cluster neural network with gradient descent technique, and improves the recognition rates. In the proposed scheme, Kirsch masks are adopted for extracting feature vectors and a three-layer cluster neural network with five independent subnetworks is developed for classifying similar numerals efficiently. In order to verify the performance of the proposed multilayer cluster neural network, experiments with handwritten numeral database of Concordia University of Canada, that of Electro-Technical Laboratory of Japan, and that of Electronics and Telecommunications Research Institute of Korea were performed. For the case of determining the initial weights using a genetic algorithm, 97.10{\%}, 99.12{\%}, and 99.40{\%} correct recognition rates were obtained, respectively.",
keywords = "Genetic algorithm, Multilayer cluster neural network, Totally unconstrained handwritten numeral recognition",
author = "Lee, {Seong Whan}",
year = "1996",
month = "12",
day = "1",
doi = "10.1109/34.506416",
language = "English",
volume = "18",
pages = "648--652",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - Off-line recognition of totally unconstrained handwritten numerals using multilayer cluster neural network

AU - Lee, Seong Whan

PY - 1996/12/1

Y1 - 1996/12/1

N2 - In this paper, we propose a new scheme for off-line recognition of totally unconstrained handwritten numerals using a simple multilayer cluster neural network trained with the back propagation algorithm and show that the use of genetic algorithms avoids the problem of finding local minima in training the multilayer cluster neural network with gradient descent technique, and improves the recognition rates. In the proposed scheme, Kirsch masks are adopted for extracting feature vectors and a three-layer cluster neural network with five independent subnetworks is developed for classifying similar numerals efficiently. In order to verify the performance of the proposed multilayer cluster neural network, experiments with handwritten numeral database of Concordia University of Canada, that of Electro-Technical Laboratory of Japan, and that of Electronics and Telecommunications Research Institute of Korea were performed. For the case of determining the initial weights using a genetic algorithm, 97.10%, 99.12%, and 99.40% correct recognition rates were obtained, respectively.

AB - In this paper, we propose a new scheme for off-line recognition of totally unconstrained handwritten numerals using a simple multilayer cluster neural network trained with the back propagation algorithm and show that the use of genetic algorithms avoids the problem of finding local minima in training the multilayer cluster neural network with gradient descent technique, and improves the recognition rates. In the proposed scheme, Kirsch masks are adopted for extracting feature vectors and a three-layer cluster neural network with five independent subnetworks is developed for classifying similar numerals efficiently. In order to verify the performance of the proposed multilayer cluster neural network, experiments with handwritten numeral database of Concordia University of Canada, that of Electro-Technical Laboratory of Japan, and that of Electronics and Telecommunications Research Institute of Korea were performed. For the case of determining the initial weights using a genetic algorithm, 97.10%, 99.12%, and 99.40% correct recognition rates were obtained, respectively.

KW - Genetic algorithm

KW - Multilayer cluster neural network

KW - Totally unconstrained handwritten numeral recognition

UR - http://www.scopus.com/inward/record.url?scp=0030173622&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030173622&partnerID=8YFLogxK

U2 - 10.1109/34.506416

DO - 10.1109/34.506416

M3 - Article

VL - 18

SP - 648

EP - 652

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 6

ER -