Two-tier architecture for unconstrained handwritten character recognition

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

In this paper, we propose an approach that combines the unsupervised and supervised learning techniques for unconstrained handwritten numeral recognition. This approach uses the Kohonen self-organizing neural network for data classification in the first stage and the learning vector quantization (LVQ) model in the second stage to improve classification accuracy. The combined architecture performs better than the Kohonen self-organizing map alone. In the proposed approach, the collection of centroids at different phases of training plays a vital role in the performance of the recognition system. Four experiments have been conducted and experimental results show that the collection of centroids in the middle of the training gives high performance in terms of speed and accuracy. The systems developed also resolve the confusion between handwritten numerals.

Original languageEnglish
Pages (from-to)585-594
Number of pages10
JournalSadhana - Academy Proceedings in Engineering Sciences
Volume27
Issue numberPART 5
Publication statusPublished - 01-12-2002

Fingerprint

Character recognition
Unsupervised learning
Vector quantization
Supervised learning
Self organizing maps
Neural networks
Experiments

All Science Journal Classification (ASJC) codes

  • General

Cite this

@article{c267414a3786427396b9e84bc14ebd6c,
title = "Two-tier architecture for unconstrained handwritten character recognition",
abstract = "In this paper, we propose an approach that combines the unsupervised and supervised learning techniques for unconstrained handwritten numeral recognition. This approach uses the Kohonen self-organizing neural network for data classification in the first stage and the learning vector quantization (LVQ) model in the second stage to improve classification accuracy. The combined architecture performs better than the Kohonen self-organizing map alone. In the proposed approach, the collection of centroids at different phases of training plays a vital role in the performance of the recognition system. Four experiments have been conducted and experimental results show that the collection of centroids in the middle of the training gives high performance in terms of speed and accuracy. The systems developed also resolve the confusion between handwritten numerals.",
author = "Prema, {K. V.} and {Subba Reddy}, {N. V.}",
year = "2002",
month = "12",
day = "1",
language = "English",
volume = "27",
pages = "585--594",
journal = "Sadhana - Academy Proceedings in Engineering Sciences",
issn = "0256-2499",
publisher = "Springer India",
number = "PART 5",

}

Two-tier architecture for unconstrained handwritten character recognition. / Prema, K. V.; Subba Reddy, N. V.

In: Sadhana - Academy Proceedings in Engineering Sciences, Vol. 27, No. PART 5, 01.12.2002, p. 585-594.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Two-tier architecture for unconstrained handwritten character recognition

AU - Prema, K. V.

AU - Subba Reddy, N. V.

PY - 2002/12/1

Y1 - 2002/12/1

N2 - In this paper, we propose an approach that combines the unsupervised and supervised learning techniques for unconstrained handwritten numeral recognition. This approach uses the Kohonen self-organizing neural network for data classification in the first stage and the learning vector quantization (LVQ) model in the second stage to improve classification accuracy. The combined architecture performs better than the Kohonen self-organizing map alone. In the proposed approach, the collection of centroids at different phases of training plays a vital role in the performance of the recognition system. Four experiments have been conducted and experimental results show that the collection of centroids in the middle of the training gives high performance in terms of speed and accuracy. The systems developed also resolve the confusion between handwritten numerals.

AB - In this paper, we propose an approach that combines the unsupervised and supervised learning techniques for unconstrained handwritten numeral recognition. This approach uses the Kohonen self-organizing neural network for data classification in the first stage and the learning vector quantization (LVQ) model in the second stage to improve classification accuracy. The combined architecture performs better than the Kohonen self-organizing map alone. In the proposed approach, the collection of centroids at different phases of training plays a vital role in the performance of the recognition system. Four experiments have been conducted and experimental results show that the collection of centroids in the middle of the training gives high performance in terms of speed and accuracy. The systems developed also resolve the confusion between handwritten numerals.

UR - http://www.scopus.com/inward/record.url?scp=0036822886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036822886&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0036822886

VL - 27

SP - 585

EP - 594

JO - Sadhana - Academy Proceedings in Engineering Sciences

JF - Sadhana - Academy Proceedings in Engineering Sciences

SN - 0256-2499

IS - PART 5

ER -