Invariant image analysis based on radon transform and svd

Osama K. Al-Shaykh, John F. Doherty

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

A Radon-based invariant image analysis method is introduced. The linearity, shift, rotation, and scaling properties of the Radon transform are utilized to achieve invariant features to translation, rotation, and scaling. The singular values of a matrix, constructed by row-stacking of projections, are used to construct the invariant feature vector. This feature vector will be used as input to a classifier, which is here, the back-propagation neural network followed by a maximum-ontput-selector. A performance function is introduced to evaluate the performance of the recognition system. This performance function can also be used to indicate how closely the pattern matches the decision template. The effectiveness of this method is illustrated by a simulation example and it is compared with the method of Zernikc moments.

Original languageEnglish (US)
Pages (from-to)123-133
Number of pages11
JournalIEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing
Volume43
Issue number2
DOIs
StatePublished - Dec 1 1996

Fingerprint

Radon
Image analysis
Method of moments
Backpropagation
Classifiers
Neural networks

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

@article{f9d6e6f4e5964ccda68d92057af8f7de,
title = "Invariant image analysis based on radon transform and svd",
abstract = "A Radon-based invariant image analysis method is introduced. The linearity, shift, rotation, and scaling properties of the Radon transform are utilized to achieve invariant features to translation, rotation, and scaling. The singular values of a matrix, constructed by row-stacking of projections, are used to construct the invariant feature vector. This feature vector will be used as input to a classifier, which is here, the back-propagation neural network followed by a maximum-ontput-selector. A performance function is introduced to evaluate the performance of the recognition system. This performance function can also be used to indicate how closely the pattern matches the decision template. The effectiveness of this method is illustrated by a simulation example and it is compared with the method of Zernikc moments.",
author = "Al-Shaykh, {Osama K.} and Doherty, {John F.}",
year = "1996",
month = "12",
day = "1",
doi = "10.1109/82.486459",
language = "English (US)",
volume = "43",
pages = "123--133",
journal = "IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing",
issn = "1057-7130",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

Invariant image analysis based on radon transform and svd. / Al-Shaykh, Osama K.; Doherty, John F.

In: IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, Vol. 43, No. 2, 01.12.1996, p. 123-133.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Invariant image analysis based on radon transform and svd

AU - Al-Shaykh, Osama K.

AU - Doherty, John F.

PY - 1996/12/1

Y1 - 1996/12/1

N2 - A Radon-based invariant image analysis method is introduced. The linearity, shift, rotation, and scaling properties of the Radon transform are utilized to achieve invariant features to translation, rotation, and scaling. The singular values of a matrix, constructed by row-stacking of projections, are used to construct the invariant feature vector. This feature vector will be used as input to a classifier, which is here, the back-propagation neural network followed by a maximum-ontput-selector. A performance function is introduced to evaluate the performance of the recognition system. This performance function can also be used to indicate how closely the pattern matches the decision template. The effectiveness of this method is illustrated by a simulation example and it is compared with the method of Zernikc moments.

AB - A Radon-based invariant image analysis method is introduced. The linearity, shift, rotation, and scaling properties of the Radon transform are utilized to achieve invariant features to translation, rotation, and scaling. The singular values of a matrix, constructed by row-stacking of projections, are used to construct the invariant feature vector. This feature vector will be used as input to a classifier, which is here, the back-propagation neural network followed by a maximum-ontput-selector. A performance function is introduced to evaluate the performance of the recognition system. This performance function can also be used to indicate how closely the pattern matches the decision template. The effectiveness of this method is illustrated by a simulation example and it is compared with the method of Zernikc moments.

UR - http://www.scopus.com/inward/record.url?scp=0030082704&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030082704&partnerID=8YFLogxK

U2 - 10.1109/82.486459

DO - 10.1109/82.486459

M3 - Article

AN - SCOPUS:0030082704

VL - 43

SP - 123

EP - 133

JO - IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing

JF - IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing

SN - 1057-7130

IS - 2

ER -