Eigenposture for classification

Nooritawati Md Tahir, Aini Hussain, Salina Abdul Samad, Hafizah Husain, Mohd. Marzuki Mustafa

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

This study outlines a mechanism for human body posture classification based on various combination of eigenspace transform which we named as 'eigenposture' using three different classifiers; the Multilayer Perceptron (MLP), Nearest Neighbour (NN) and Probabilistic Neural Network (PNN). We apply principal component transformation to extract the features from human shape silhouettes. A combination of them was used to classify the posture of standing and non standing based on the human shape obtained from segmentation process. Different classifiers are compared to each other with respect to classification performance. Results show that combination of second and fourth eigenpostures outperformed the other eigenpostures combination.

Original languageEnglish
Pages (from-to)419-424
Number of pages6
JournalJournal of Applied Sciences
Volume6
Issue number2
DOIs
Publication statusPublished - Feb 2006

Fingerprint

Classifiers
Multilayer neural networks
Neural networks

Keywords

  • Artificial neural network
  • Classifier
  • Nearest neighbour
  • Principal component analysis
  • Silhouette

ASJC Scopus subject areas

  • General

Cite this

Eigenposture for classification. / Tahir, Nooritawati Md; Hussain, Aini; Abdul Samad, Salina; Husain, Hafizah; Mustafa, Mohd. Marzuki.

In: Journal of Applied Sciences, Vol. 6, No. 2, 02.2006, p. 419-424.

Research output: Contribution to journalArticle

Tahir, Nooritawati Md ; Hussain, Aini ; Abdul Samad, Salina ; Husain, Hafizah ; Mustafa, Mohd. Marzuki. / Eigenposture for classification. In: Journal of Applied Sciences. 2006 ; Vol. 6, No. 2. pp. 419-424.
@article{289be8c5a1db4ccebbe34b271b4d6639,
title = "Eigenposture for classification",
abstract = "This study outlines a mechanism for human body posture classification based on various combination of eigenspace transform which we named as 'eigenposture' using three different classifiers; the Multilayer Perceptron (MLP), Nearest Neighbour (NN) and Probabilistic Neural Network (PNN). We apply principal component transformation to extract the features from human shape silhouettes. A combination of them was used to classify the posture of standing and non standing based on the human shape obtained from segmentation process. Different classifiers are compared to each other with respect to classification performance. Results show that combination of second and fourth eigenpostures outperformed the other eigenpostures combination.",
keywords = "Artificial neural network, Classifier, Nearest neighbour, Principal component analysis, Silhouette",
author = "Tahir, {Nooritawati Md} and Aini Hussain and {Abdul Samad}, Salina and Hafizah Husain and Mustafa, {Mohd. Marzuki}",
year = "2006",
month = "2",
doi = "10.3923/jas.2006.419.424",
language = "English",
volume = "6",
pages = "419--424",
journal = "Journal of Applied Sciences",
issn = "1812-5654",
publisher = "Asian Network for Scientific Information",
number = "2",

}

TY - JOUR

T1 - Eigenposture for classification

AU - Tahir, Nooritawati Md

AU - Hussain, Aini

AU - Abdul Samad, Salina

AU - Husain, Hafizah

AU - Mustafa, Mohd. Marzuki

PY - 2006/2

Y1 - 2006/2

N2 - This study outlines a mechanism for human body posture classification based on various combination of eigenspace transform which we named as 'eigenposture' using three different classifiers; the Multilayer Perceptron (MLP), Nearest Neighbour (NN) and Probabilistic Neural Network (PNN). We apply principal component transformation to extract the features from human shape silhouettes. A combination of them was used to classify the posture of standing and non standing based on the human shape obtained from segmentation process. Different classifiers are compared to each other with respect to classification performance. Results show that combination of second and fourth eigenpostures outperformed the other eigenpostures combination.

AB - This study outlines a mechanism for human body posture classification based on various combination of eigenspace transform which we named as 'eigenposture' using three different classifiers; the Multilayer Perceptron (MLP), Nearest Neighbour (NN) and Probabilistic Neural Network (PNN). We apply principal component transformation to extract the features from human shape silhouettes. A combination of them was used to classify the posture of standing and non standing based on the human shape obtained from segmentation process. Different classifiers are compared to each other with respect to classification performance. Results show that combination of second and fourth eigenpostures outperformed the other eigenpostures combination.

KW - Artificial neural network

KW - Classifier

KW - Nearest neighbour

KW - Principal component analysis

KW - Silhouette

UR - http://www.scopus.com/inward/record.url?scp=33644993394&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33644993394&partnerID=8YFLogxK

U2 - 10.3923/jas.2006.419.424

DO - 10.3923/jas.2006.419.424

M3 - Article

AN - SCOPUS:33644993394

VL - 6

SP - 419

EP - 424

JO - Journal of Applied Sciences

JF - Journal of Applied Sciences

SN - 1812-5654

IS - 2

ER -