Integrated segmentation and recognition of connected handwritten characters with recurrent neural network

Seong Whan Lee, Eung Jae Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

In this paper, we propose an efficient method for integrated segmentation and recognition of connected handwritten characters with recurrent neural network. In the proposed method, a new type of recurrent neural network is developed for training the spatial dependencies in connected handwritten characters. This recurrent neural network differs from Jordan's and Elman's recurrent networks in view of functions and architectures because it was originally extended from the multilayer feedforward neural network for improving the discrimination and generalization power. In order to verify the performance of the proposed method, experiments with the NIST database have been performed and the performance of the proposed method has been compared with those of the previous integrated segmentation and recognition methods. The experimental results reveal that the proposed method is superior to the previous integrated segmentation and recognition methods in view of discrimination and generalization ability.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsLuc M. Vincent, Jonathan J. Hull
Pages251-261
Number of pages11
Volume2660
Publication statusPublished - 1996 Jan 1
EventDocument Recognition III - San Jose, CA, USA
Duration: 1996 Jan 291996 Jan 30

Other

OtherDocument Recognition III
CitySan Jose, CA, USA
Period96/1/2996/1/30

Fingerprint

Recurrent neural networks
Feedforward neural networks
Multilayer neural networks
discrimination
spatial dependencies
Jordan
education
Experiments

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Lee, S. W., & Lee, E. J. (1996). Integrated segmentation and recognition of connected handwritten characters with recurrent neural network. In L. M. Vincent, & J. J. Hull (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 2660, pp. 251-261)

Integrated segmentation and recognition of connected handwritten characters with recurrent neural network. / Lee, Seong Whan; Lee, Eung Jae.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / Luc M. Vincent; Jonathan J. Hull. Vol. 2660 1996. p. 251-261.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lee, SW & Lee, EJ 1996, Integrated segmentation and recognition of connected handwritten characters with recurrent neural network. in LM Vincent & JJ Hull (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 2660, pp. 251-261, Document Recognition III, San Jose, CA, USA, 96/1/29.
Lee SW, Lee EJ. Integrated segmentation and recognition of connected handwritten characters with recurrent neural network. In Vincent LM, Hull JJ, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 2660. 1996. p. 251-261
Lee, Seong Whan ; Lee, Eung Jae. / Integrated segmentation and recognition of connected handwritten characters with recurrent neural network. Proceedings of SPIE - The International Society for Optical Engineering. editor / Luc M. Vincent ; Jonathan J. Hull. Vol. 2660 1996. pp. 251-261
@inproceedings{569dd997717243dc91e02c08f2b3bb3d,
title = "Integrated segmentation and recognition of connected handwritten characters with recurrent neural network",
abstract = "In this paper, we propose an efficient method for integrated segmentation and recognition of connected handwritten characters with recurrent neural network. In the proposed method, a new type of recurrent neural network is developed for training the spatial dependencies in connected handwritten characters. This recurrent neural network differs from Jordan's and Elman's recurrent networks in view of functions and architectures because it was originally extended from the multilayer feedforward neural network for improving the discrimination and generalization power. In order to verify the performance of the proposed method, experiments with the NIST database have been performed and the performance of the proposed method has been compared with those of the previous integrated segmentation and recognition methods. The experimental results reveal that the proposed method is superior to the previous integrated segmentation and recognition methods in view of discrimination and generalization ability.",
author = "Lee, {Seong Whan} and Lee, {Eung Jae}",
year = "1996",
month = "1",
day = "1",
language = "English",
isbn = "0819420344",
volume = "2660",
pages = "251--261",
editor = "Vincent, {Luc M.} and Hull, {Jonathan J.}",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Integrated segmentation and recognition of connected handwritten characters with recurrent neural network

AU - Lee, Seong Whan

AU - Lee, Eung Jae

PY - 1996/1/1

Y1 - 1996/1/1

N2 - In this paper, we propose an efficient method for integrated segmentation and recognition of connected handwritten characters with recurrent neural network. In the proposed method, a new type of recurrent neural network is developed for training the spatial dependencies in connected handwritten characters. This recurrent neural network differs from Jordan's and Elman's recurrent networks in view of functions and architectures because it was originally extended from the multilayer feedforward neural network for improving the discrimination and generalization power. In order to verify the performance of the proposed method, experiments with the NIST database have been performed and the performance of the proposed method has been compared with those of the previous integrated segmentation and recognition methods. The experimental results reveal that the proposed method is superior to the previous integrated segmentation and recognition methods in view of discrimination and generalization ability.

AB - In this paper, we propose an efficient method for integrated segmentation and recognition of connected handwritten characters with recurrent neural network. In the proposed method, a new type of recurrent neural network is developed for training the spatial dependencies in connected handwritten characters. This recurrent neural network differs from Jordan's and Elman's recurrent networks in view of functions and architectures because it was originally extended from the multilayer feedforward neural network for improving the discrimination and generalization power. In order to verify the performance of the proposed method, experiments with the NIST database have been performed and the performance of the proposed method has been compared with those of the previous integrated segmentation and recognition methods. The experimental results reveal that the proposed method is superior to the previous integrated segmentation and recognition methods in view of discrimination and generalization ability.

UR - http://www.scopus.com/inward/record.url?scp=0029756109&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029756109&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0029756109

SN - 0819420344

SN - 9780819420343

VL - 2660

SP - 251

EP - 261

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

A2 - Vincent, Luc M.

A2 - Hull, Jonathan J.

ER -