Analysis of PCA based feature vectors for SVM posture classification

Shahrani Shahbudin, Aini Hussain, Hafizah Husain, Salina Abdul Samad, Nooritawati Md Tahir

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

7 Citations (Scopus)

Abstract

Many classifiers have been employed to classify human posture classification; however, most of them only presents the average accuracy of the classification. Furthermore, the details of the measured parameters especially for SVM classifier are not measured. Therefore, the objective of this work is to analyse and classify human body posture using Support Vector Machine (SVM) techniques based on various two combinations of eigenpostures by considering two different solvers in the training process. The two solvers namely Sequential Minimal Optimization (SMO) and Matlab Quadratics Programming (QP) solvers have been studied and analyzed to perform the SVM training. The principal component analysis (PCA) method is applied to extract the features from human shape silhouettes. These extracted feature vectors are then used to perform human posture classification. Human posture evaluates which eigenpostures (feature vectors of the several eigenvalues) can be used to classify either human standing posture or human non-standing posture. Next, the solvers that produced the best performance in classifying human postures as well as the best combination of eigenpostures were selected. The results verified that the combination of second and fourth eigenpostures gives the superb performance with 100% correct classification and it is shown that the best solver in training process to classify human body posture classification is the SMO based on the shortest CPU time attained.

Original languageEnglish
Title of host publicationProceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications
DOIs
Publication statusPublished - 2010
Event2010 6th International Colloquium on Signal Processing and Its Applications, CSPA 2010 - Melaka
Duration: 21 May 201023 May 2010

Other

Other2010 6th International Colloquium on Signal Processing and Its Applications, CSPA 2010
CityMelaka
Period21/5/1023/5/10

Fingerprint

Principal component analysis
Support vector machines
Classifiers
Quadratic programming
Program processors

Keywords

  • Decision boundary
  • Eigenpostures
  • Matlab Quadratics Programming (QP) solver
  • Sequential Minimal Optimization (SMO)
  • Support Vector Machines (SVM)

ASJC Scopus subject areas

  • Artificial Intelligence
  • Signal Processing
  • Control and Systems Engineering

Cite this

Shahbudin, S., Hussain, A., Husain, H., Abdul Samad, S., & Tahir, N. M. (2010). Analysis of PCA based feature vectors for SVM posture classification. In Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications [5545268] https://doi.org/10.1109/CSPA.2010.5545268

Analysis of PCA based feature vectors for SVM posture classification. / Shahbudin, Shahrani; Hussain, Aini; Husain, Hafizah; Abdul Samad, Salina; Tahir, Nooritawati Md.

Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications. 2010. 5545268.

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

Shahbudin, S, Hussain, A, Husain, H, Abdul Samad, S & Tahir, NM 2010, Analysis of PCA based feature vectors for SVM posture classification. in Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications., 5545268, 2010 6th International Colloquium on Signal Processing and Its Applications, CSPA 2010, Melaka, 21/5/10. https://doi.org/10.1109/CSPA.2010.5545268
Shahbudin S, Hussain A, Husain H, Abdul Samad S, Tahir NM. Analysis of PCA based feature vectors for SVM posture classification. In Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications. 2010. 5545268 https://doi.org/10.1109/CSPA.2010.5545268
Shahbudin, Shahrani ; Hussain, Aini ; Husain, Hafizah ; Abdul Samad, Salina ; Tahir, Nooritawati Md. / Analysis of PCA based feature vectors for SVM posture classification. Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications. 2010.
@inproceedings{e72bc2c576324ef6af1a6eed046ad346,
title = "Analysis of PCA based feature vectors for SVM posture classification",
abstract = "Many classifiers have been employed to classify human posture classification; however, most of them only presents the average accuracy of the classification. Furthermore, the details of the measured parameters especially for SVM classifier are not measured. Therefore, the objective of this work is to analyse and classify human body posture using Support Vector Machine (SVM) techniques based on various two combinations of eigenpostures by considering two different solvers in the training process. The two solvers namely Sequential Minimal Optimization (SMO) and Matlab Quadratics Programming (QP) solvers have been studied and analyzed to perform the SVM training. The principal component analysis (PCA) method is applied to extract the features from human shape silhouettes. These extracted feature vectors are then used to perform human posture classification. Human posture evaluates which eigenpostures (feature vectors of the several eigenvalues) can be used to classify either human standing posture or human non-standing posture. Next, the solvers that produced the best performance in classifying human postures as well as the best combination of eigenpostures were selected. The results verified that the combination of second and fourth eigenpostures gives the superb performance with 100{\%} correct classification and it is shown that the best solver in training process to classify human body posture classification is the SMO based on the shortest CPU time attained.",
keywords = "Decision boundary, Eigenpostures, Matlab Quadratics Programming (QP) solver, Sequential Minimal Optimization (SMO), Support Vector Machines (SVM)",
author = "Shahrani Shahbudin and Aini Hussain and Hafizah Husain and {Abdul Samad}, Salina and Tahir, {Nooritawati Md}",
year = "2010",
doi = "10.1109/CSPA.2010.5545268",
language = "English",
isbn = "9781424471225",
booktitle = "Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications",

}

TY - GEN

T1 - Analysis of PCA based feature vectors for SVM posture classification

AU - Shahbudin, Shahrani

AU - Hussain, Aini

AU - Husain, Hafizah

AU - Abdul Samad, Salina

AU - Tahir, Nooritawati Md

PY - 2010

Y1 - 2010

N2 - Many classifiers have been employed to classify human posture classification; however, most of them only presents the average accuracy of the classification. Furthermore, the details of the measured parameters especially for SVM classifier are not measured. Therefore, the objective of this work is to analyse and classify human body posture using Support Vector Machine (SVM) techniques based on various two combinations of eigenpostures by considering two different solvers in the training process. The two solvers namely Sequential Minimal Optimization (SMO) and Matlab Quadratics Programming (QP) solvers have been studied and analyzed to perform the SVM training. The principal component analysis (PCA) method is applied to extract the features from human shape silhouettes. These extracted feature vectors are then used to perform human posture classification. Human posture evaluates which eigenpostures (feature vectors of the several eigenvalues) can be used to classify either human standing posture or human non-standing posture. Next, the solvers that produced the best performance in classifying human postures as well as the best combination of eigenpostures were selected. The results verified that the combination of second and fourth eigenpostures gives the superb performance with 100% correct classification and it is shown that the best solver in training process to classify human body posture classification is the SMO based on the shortest CPU time attained.

AB - Many classifiers have been employed to classify human posture classification; however, most of them only presents the average accuracy of the classification. Furthermore, the details of the measured parameters especially for SVM classifier are not measured. Therefore, the objective of this work is to analyse and classify human body posture using Support Vector Machine (SVM) techniques based on various two combinations of eigenpostures by considering two different solvers in the training process. The two solvers namely Sequential Minimal Optimization (SMO) and Matlab Quadratics Programming (QP) solvers have been studied and analyzed to perform the SVM training. The principal component analysis (PCA) method is applied to extract the features from human shape silhouettes. These extracted feature vectors are then used to perform human posture classification. Human posture evaluates which eigenpostures (feature vectors of the several eigenvalues) can be used to classify either human standing posture or human non-standing posture. Next, the solvers that produced the best performance in classifying human postures as well as the best combination of eigenpostures were selected. The results verified that the combination of second and fourth eigenpostures gives the superb performance with 100% correct classification and it is shown that the best solver in training process to classify human body posture classification is the SMO based on the shortest CPU time attained.

KW - Decision boundary

KW - Eigenpostures

KW - Matlab Quadratics Programming (QP) solver

KW - Sequential Minimal Optimization (SMO)

KW - Support Vector Machines (SVM)

UR - http://www.scopus.com/inward/record.url?scp=77956517406&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956517406&partnerID=8YFLogxK

U2 - 10.1109/CSPA.2010.5545268

DO - 10.1109/CSPA.2010.5545268

M3 - Conference contribution

SN - 9781424471225

BT - Proceedings - CSPA 2010: 2010 6th International Colloquium on Signal Processing and Its Applications

ER -