A rotation, scaling, and translation invariant pattern classification system

Cem Yüceer, Kemal Oflazer

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

This paper describes a hybrid pattern classification system based on a pattern preprocessor and an artificial neural network classifier that can recognize patterns even when they are deformed by transformation of rotation, scaling, and translation or a combination of these. After a description of the system architecture we provide experimental results from three different classification domains: classification of letters in the English alphabet, classification of the letters in the Japanese Katakana alphabet, and classification of geometric figures. For the first problem, our system can recognize patterns deformed by a single transformation with well over 90% success ratio and with 89% success ratio when all three transformations are applied. For the second problem, the system performs very good for patterns deformed by scaling and translation but worse (about 75%) when rotations are involved. For the third problem, the success ratio is almost 100% when only a single transformation is applied and 88% when all three transformations are applied. The system is general purpose and has a reasonable noise tolerance.

Original languageEnglish
Pages (from-to)687-710
Number of pages24
JournalPattern Recognition
Volume26
Issue number5
DOIs
Publication statusPublished - 1993
Externally publishedYes

Fingerprint

Pattern recognition
Computer systems
Classifiers
Neural networks

Keywords

  • Artificial neural networks
  • Deformation invariant pattern classification
  • Pattern recognition

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

A rotation, scaling, and translation invariant pattern classification system. / Yüceer, Cem; Oflazer, Kemal.

In: Pattern Recognition, Vol. 26, No. 5, 1993, p. 687-710.

Research output: Contribution to journalArticle

Yüceer, Cem ; Oflazer, Kemal. / A rotation, scaling, and translation invariant pattern classification system. In: Pattern Recognition. 1993 ; Vol. 26, No. 5. pp. 687-710.
@article{52e82928909f4fa285de5d0fe9e4f017,
title = "A rotation, scaling, and translation invariant pattern classification system",
abstract = "This paper describes a hybrid pattern classification system based on a pattern preprocessor and an artificial neural network classifier that can recognize patterns even when they are deformed by transformation of rotation, scaling, and translation or a combination of these. After a description of the system architecture we provide experimental results from three different classification domains: classification of letters in the English alphabet, classification of the letters in the Japanese Katakana alphabet, and classification of geometric figures. For the first problem, our system can recognize patterns deformed by a single transformation with well over 90{\%} success ratio and with 89{\%} success ratio when all three transformations are applied. For the second problem, the system performs very good for patterns deformed by scaling and translation but worse (about 75{\%}) when rotations are involved. For the third problem, the success ratio is almost 100{\%} when only a single transformation is applied and 88{\%} when all three transformations are applied. The system is general purpose and has a reasonable noise tolerance.",
keywords = "Artificial neural networks, Deformation invariant pattern classification, Pattern recognition",
author = "Cem Y{\"u}ceer and Kemal Oflazer",
year = "1993",
doi = "10.1016/0031-3203(93)90122-D",
language = "English",
volume = "26",
pages = "687--710",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "5",

}

TY - JOUR

T1 - A rotation, scaling, and translation invariant pattern classification system

AU - Yüceer, Cem

AU - Oflazer, Kemal

PY - 1993

Y1 - 1993

N2 - This paper describes a hybrid pattern classification system based on a pattern preprocessor and an artificial neural network classifier that can recognize patterns even when they are deformed by transformation of rotation, scaling, and translation or a combination of these. After a description of the system architecture we provide experimental results from three different classification domains: classification of letters in the English alphabet, classification of the letters in the Japanese Katakana alphabet, and classification of geometric figures. For the first problem, our system can recognize patterns deformed by a single transformation with well over 90% success ratio and with 89% success ratio when all three transformations are applied. For the second problem, the system performs very good for patterns deformed by scaling and translation but worse (about 75%) when rotations are involved. For the third problem, the success ratio is almost 100% when only a single transformation is applied and 88% when all three transformations are applied. The system is general purpose and has a reasonable noise tolerance.

AB - This paper describes a hybrid pattern classification system based on a pattern preprocessor and an artificial neural network classifier that can recognize patterns even when they are deformed by transformation of rotation, scaling, and translation or a combination of these. After a description of the system architecture we provide experimental results from three different classification domains: classification of letters in the English alphabet, classification of the letters in the Japanese Katakana alphabet, and classification of geometric figures. For the first problem, our system can recognize patterns deformed by a single transformation with well over 90% success ratio and with 89% success ratio when all three transformations are applied. For the second problem, the system performs very good for patterns deformed by scaling and translation but worse (about 75%) when rotations are involved. For the third problem, the success ratio is almost 100% when only a single transformation is applied and 88% when all three transformations are applied. The system is general purpose and has a reasonable noise tolerance.

KW - Artificial neural networks

KW - Deformation invariant pattern classification

KW - Pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=0027595913&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027595913&partnerID=8YFLogxK

U2 - 10.1016/0031-3203(93)90122-D

DO - 10.1016/0031-3203(93)90122-D

M3 - Article

AN - SCOPUS:0027595913

VL - 26

SP - 687

EP - 710

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 5

ER -