Feature extraction with an improved scale-invariants for deformation digits

S. M. Shamsuddin, M. N. Sulaiman, Maslina Darus

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

This paper presents an alternative formulation on improved scaled-invariants using higher order centralized moments for digits with deformations. We claim that deformation digits would be digits with improper shapes, unconstrained styles of writing and different orientations. A detail experimental evaluation of the utilizing various moments order as pattern features in recognition of handprinted and handwritten digits have been carried out using the proposed invariants. We use scale-invariants of centralized moments of order 2 for the numerator and order 4 for the denominator while preserving the scale factor of the same order. Unconstrained digits are rotated clockwise and counter clockwise of 45 degree. As a comparison, we generate geometric moment invariants on these digits. We train these invariants using standard back-propagation and modified backpropagation in the classifications phase. We found that the results are promising with an improved scaled-invariants of higher order, and the classifications of the digits are successfully recognized.

Original languageEnglish
Pages (from-to)13-23
Number of pages11
JournalInternational Journal of Computer Mathematics
Volume76
Issue number1
Publication statusPublished - 2000

Fingerprint

Scale Invariant
Backpropagation
Digit
Feature Extraction
Feature extraction
Invariant
Back Propagation
Moment
Higher Order
Moment Invariants
Anticlockwise
Clockwise
Geometric Invariants
Numerator
Scale factor
Denominator
Experimental Evaluation
Formulation
Alternatives

ASJC Scopus subject areas

  • Applied Mathematics

Cite this

Feature extraction with an improved scale-invariants for deformation digits. / Shamsuddin, S. M.; Sulaiman, M. N.; Darus, Maslina.

In: International Journal of Computer Mathematics, Vol. 76, No. 1, 2000, p. 13-23.

Research output: Contribution to journalArticle

@article{0745e6fea4c44720951cf14821822210,
title = "Feature extraction with an improved scale-invariants for deformation digits",
abstract = "This paper presents an alternative formulation on improved scaled-invariants using higher order centralized moments for digits with deformations. We claim that deformation digits would be digits with improper shapes, unconstrained styles of writing and different orientations. A detail experimental evaluation of the utilizing various moments order as pattern features in recognition of handprinted and handwritten digits have been carried out using the proposed invariants. We use scale-invariants of centralized moments of order 2 for the numerator and order 4 for the denominator while preserving the scale factor of the same order. Unconstrained digits are rotated clockwise and counter clockwise of 45 degree. As a comparison, we generate geometric moment invariants on these digits. We train these invariants using standard back-propagation and modified backpropagation in the classifications phase. We found that the results are promising with an improved scaled-invariants of higher order, and the classifications of the digits are successfully recognized.",
author = "Shamsuddin, {S. M.} and Sulaiman, {M. N.} and Maslina Darus",
year = "2000",
language = "English",
volume = "76",
pages = "13--23",
journal = "International Journal of Computer Mathematics",
issn = "0020-7160",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

TY - JOUR

T1 - Feature extraction with an improved scale-invariants for deformation digits

AU - Shamsuddin, S. M.

AU - Sulaiman, M. N.

AU - Darus, Maslina

PY - 2000

Y1 - 2000

N2 - This paper presents an alternative formulation on improved scaled-invariants using higher order centralized moments for digits with deformations. We claim that deformation digits would be digits with improper shapes, unconstrained styles of writing and different orientations. A detail experimental evaluation of the utilizing various moments order as pattern features in recognition of handprinted and handwritten digits have been carried out using the proposed invariants. We use scale-invariants of centralized moments of order 2 for the numerator and order 4 for the denominator while preserving the scale factor of the same order. Unconstrained digits are rotated clockwise and counter clockwise of 45 degree. As a comparison, we generate geometric moment invariants on these digits. We train these invariants using standard back-propagation and modified backpropagation in the classifications phase. We found that the results are promising with an improved scaled-invariants of higher order, and the classifications of the digits are successfully recognized.

AB - This paper presents an alternative formulation on improved scaled-invariants using higher order centralized moments for digits with deformations. We claim that deformation digits would be digits with improper shapes, unconstrained styles of writing and different orientations. A detail experimental evaluation of the utilizing various moments order as pattern features in recognition of handprinted and handwritten digits have been carried out using the proposed invariants. We use scale-invariants of centralized moments of order 2 for the numerator and order 4 for the denominator while preserving the scale factor of the same order. Unconstrained digits are rotated clockwise and counter clockwise of 45 degree. As a comparison, we generate geometric moment invariants on these digits. We train these invariants using standard back-propagation and modified backpropagation in the classifications phase. We found that the results are promising with an improved scaled-invariants of higher order, and the classifications of the digits are successfully recognized.

UR - http://www.scopus.com/inward/record.url?scp=0034479334&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034479334&partnerID=8YFLogxK

M3 - Article

VL - 76

SP - 13

EP - 23

JO - International Journal of Computer Mathematics

JF - International Journal of Computer Mathematics

SN - 0020-7160

IS - 1

ER -