Handwritten character recognition using a gradient classifier

Robert D. Brandt, Yao Wang, Alan J. Laub, Sanjit K. Mitra

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

Abstract

The authors consider a prototype-based character recognizer that makes comparisons based on blurred representations of the images. The blurring induces a metric on the space of all images that varies continuously under continuous deformations of the image plane. This blurred representation is suitable for direct implementation of a nearest neighbor classifer. However, it is still desirable to have a representation which is invariant under rotation, translation, and scaling of the image plane. A representation which is locally invariant under these transformations is produced by transforming an input to a local minimum of its distance from each prototype simultaneously. These minima are found by performing a gradient descent on an appropriate error surface over the four transformation parameters. The error functional is the L2-norm of the difference between the blurred prototype and the blurred input. The resulting classifier makes more efficient use of prototypes than the nearest neighbor classifier.

Original languageEnglish (US)
Title of host publicationProc 1988 IEEE Int Conf Syst Man Cybern
Editors Anon
Pages361-364
Number of pages4
StatePublished - 1988
EventProceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics - Beijing/Shenyang, China
Duration: Aug 8 1988Aug 12 1988

Other

OtherProceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics
CityBeijing/Shenyang, China
Period8/8/888/12/88

Fingerprint

Character recognition
Classifiers

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Brandt, R. D., Wang, Y., Laub, A. J., & Mitra, S. K. (1988). Handwritten character recognition using a gradient classifier. In Anon (Ed.), Proc 1988 IEEE Int Conf Syst Man Cybern (pp. 361-364)

Handwritten character recognition using a gradient classifier. / Brandt, Robert D.; Wang, Yao; Laub, Alan J.; Mitra, Sanjit K.

Proc 1988 IEEE Int Conf Syst Man Cybern. ed. / Anon. 1988. p. 361-364.

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

Brandt, RD, Wang, Y, Laub, AJ & Mitra, SK 1988, Handwritten character recognition using a gradient classifier. in Anon (ed.), Proc 1988 IEEE Int Conf Syst Man Cybern. pp. 361-364, Proceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics, Beijing/Shenyang, China, 8/8/88.
Brandt RD, Wang Y, Laub AJ, Mitra SK. Handwritten character recognition using a gradient classifier. In Anon, editor, Proc 1988 IEEE Int Conf Syst Man Cybern. 1988. p. 361-364
Brandt, Robert D. ; Wang, Yao ; Laub, Alan J. ; Mitra, Sanjit K. / Handwritten character recognition using a gradient classifier. Proc 1988 IEEE Int Conf Syst Man Cybern. editor / Anon. 1988. pp. 361-364
@inproceedings{f90c714d006b453ebcc6e76cf1f83389,
title = "Handwritten character recognition using a gradient classifier",
abstract = "The authors consider a prototype-based character recognizer that makes comparisons based on blurred representations of the images. The blurring induces a metric on the space of all images that varies continuously under continuous deformations of the image plane. This blurred representation is suitable for direct implementation of a nearest neighbor classifer. However, it is still desirable to have a representation which is invariant under rotation, translation, and scaling of the image plane. A representation which is locally invariant under these transformations is produced by transforming an input to a local minimum of its distance from each prototype simultaneously. These minima are found by performing a gradient descent on an appropriate error surface over the four transformation parameters. The error functional is the L2-norm of the difference between the blurred prototype and the blurred input. The resulting classifier makes more efficient use of prototypes than the nearest neighbor classifier.",
author = "Brandt, {Robert D.} and Yao Wang and Laub, {Alan J.} and Mitra, {Sanjit K.}",
year = "1988",
language = "English (US)",
isbn = "7800030393",
pages = "361--364",
editor = "Anon",
booktitle = "Proc 1988 IEEE Int Conf Syst Man Cybern",

}

TY - GEN

T1 - Handwritten character recognition using a gradient classifier

AU - Brandt, Robert D.

AU - Wang, Yao

AU - Laub, Alan J.

AU - Mitra, Sanjit K.

PY - 1988

Y1 - 1988

N2 - The authors consider a prototype-based character recognizer that makes comparisons based on blurred representations of the images. The blurring induces a metric on the space of all images that varies continuously under continuous deformations of the image plane. This blurred representation is suitable for direct implementation of a nearest neighbor classifer. However, it is still desirable to have a representation which is invariant under rotation, translation, and scaling of the image plane. A representation which is locally invariant under these transformations is produced by transforming an input to a local minimum of its distance from each prototype simultaneously. These minima are found by performing a gradient descent on an appropriate error surface over the four transformation parameters. The error functional is the L2-norm of the difference between the blurred prototype and the blurred input. The resulting classifier makes more efficient use of prototypes than the nearest neighbor classifier.

AB - The authors consider a prototype-based character recognizer that makes comparisons based on blurred representations of the images. The blurring induces a metric on the space of all images that varies continuously under continuous deformations of the image plane. This blurred representation is suitable for direct implementation of a nearest neighbor classifer. However, it is still desirable to have a representation which is invariant under rotation, translation, and scaling of the image plane. A representation which is locally invariant under these transformations is produced by transforming an input to a local minimum of its distance from each prototype simultaneously. These minima are found by performing a gradient descent on an appropriate error surface over the four transformation parameters. The error functional is the L2-norm of the difference between the blurred prototype and the blurred input. The resulting classifier makes more efficient use of prototypes than the nearest neighbor classifier.

UR - http://www.scopus.com/inward/record.url?scp=0024169104&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024169104&partnerID=8YFLogxK

M3 - Conference contribution

SN - 7800030393

SP - 361

EP - 364

BT - Proc 1988 IEEE Int Conf Syst Man Cybern

A2 - Anon, null

ER -