Model-based gait analysis for gender recognition

Ahmad Puad Ismail, Nooritawati Md Tahir, Aini Hussain

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

3 Citations (Scopus)

Abstract

Gender classification via model-based human gait data is still immature. Hence in this research, the possibility of side view human gait silhouette to be used as gender recognition is evaluated using model-based approach. Firstly, six attributes located at lower part of human gait specifically from below waist onwards have been identified as the significant points are skeletonized based on the human gait silhouette attained. Next the vertical angles of both hip and knee with respect to thigh for 32 image sequences are extracted as feature vectors followed by feature selection via statistical analysis specifically analysis of variance along with multiple comparison procedure. Further, the resultant of feature selection acted as inputs to the artificial neural network classifier. Initial findings with accuracy of 90% and above confirmed that the proposed method suited to be utilized as gender recognition based on human gait.

Original languageEnglish
Title of host publicationProceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012
Pages400-403
Number of pages4
DOIs
Publication statusPublished - 2012
Event2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012 - Melaka
Duration: 23 Mar 201225 Mar 2012

Other

Other2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012
CityMelaka
Period23/3/1225/3/12

Fingerprint

Gait analysis
Feature extraction
Analysis of variance (ANOVA)
Statistical methods
Classifiers
Neural networks

Keywords

  • ANOVA
  • artificial neural network
  • gender classification
  • human gait
  • Multiple Comparison Procedure

ASJC Scopus subject areas

  • Signal Processing

Cite this

Ismail, A. P., Tahir, N. M., & Hussain, A. (2012). Model-based gait analysis for gender recognition. In Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012 (pp. 400-403). [6194757] https://doi.org/10.1109/CSPA.2012.6194757

Model-based gait analysis for gender recognition. / Ismail, Ahmad Puad; Tahir, Nooritawati Md; Hussain, Aini.

Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012. 2012. p. 400-403 6194757.

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

Ismail, AP, Tahir, NM & Hussain, A 2012, Model-based gait analysis for gender recognition. in Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012., 6194757, pp. 400-403, 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012, Melaka, 23/3/12. https://doi.org/10.1109/CSPA.2012.6194757
Ismail AP, Tahir NM, Hussain A. Model-based gait analysis for gender recognition. In Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012. 2012. p. 400-403. 6194757 https://doi.org/10.1109/CSPA.2012.6194757
Ismail, Ahmad Puad ; Tahir, Nooritawati Md ; Hussain, Aini. / Model-based gait analysis for gender recognition. Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012. 2012. pp. 400-403
@inproceedings{e3a952ff413740ec9d05439d03e9a0f3,
title = "Model-based gait analysis for gender recognition",
abstract = "Gender classification via model-based human gait data is still immature. Hence in this research, the possibility of side view human gait silhouette to be used as gender recognition is evaluated using model-based approach. Firstly, six attributes located at lower part of human gait specifically from below waist onwards have been identified as the significant points are skeletonized based on the human gait silhouette attained. Next the vertical angles of both hip and knee with respect to thigh for 32 image sequences are extracted as feature vectors followed by feature selection via statistical analysis specifically analysis of variance along with multiple comparison procedure. Further, the resultant of feature selection acted as inputs to the artificial neural network classifier. Initial findings with accuracy of 90{\%} and above confirmed that the proposed method suited to be utilized as gender recognition based on human gait.",
keywords = "ANOVA, artificial neural network, gender classification, human gait, Multiple Comparison Procedure",
author = "Ismail, {Ahmad Puad} and Tahir, {Nooritawati Md} and Aini Hussain",
year = "2012",
doi = "10.1109/CSPA.2012.6194757",
language = "English",
isbn = "9781467309615",
pages = "400--403",
booktitle = "Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012",

}

TY - GEN

T1 - Model-based gait analysis for gender recognition

AU - Ismail, Ahmad Puad

AU - Tahir, Nooritawati Md

AU - Hussain, Aini

PY - 2012

Y1 - 2012

N2 - Gender classification via model-based human gait data is still immature. Hence in this research, the possibility of side view human gait silhouette to be used as gender recognition is evaluated using model-based approach. Firstly, six attributes located at lower part of human gait specifically from below waist onwards have been identified as the significant points are skeletonized based on the human gait silhouette attained. Next the vertical angles of both hip and knee with respect to thigh for 32 image sequences are extracted as feature vectors followed by feature selection via statistical analysis specifically analysis of variance along with multiple comparison procedure. Further, the resultant of feature selection acted as inputs to the artificial neural network classifier. Initial findings with accuracy of 90% and above confirmed that the proposed method suited to be utilized as gender recognition based on human gait.

AB - Gender classification via model-based human gait data is still immature. Hence in this research, the possibility of side view human gait silhouette to be used as gender recognition is evaluated using model-based approach. Firstly, six attributes located at lower part of human gait specifically from below waist onwards have been identified as the significant points are skeletonized based on the human gait silhouette attained. Next the vertical angles of both hip and knee with respect to thigh for 32 image sequences are extracted as feature vectors followed by feature selection via statistical analysis specifically analysis of variance along with multiple comparison procedure. Further, the resultant of feature selection acted as inputs to the artificial neural network classifier. Initial findings with accuracy of 90% and above confirmed that the proposed method suited to be utilized as gender recognition based on human gait.

KW - ANOVA

KW - artificial neural network

KW - gender classification

KW - human gait

KW - Multiple Comparison Procedure

UR - http://www.scopus.com/inward/record.url?scp=84861539512&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861539512&partnerID=8YFLogxK

U2 - 10.1109/CSPA.2012.6194757

DO - 10.1109/CSPA.2012.6194757

M3 - Conference contribution

AN - SCOPUS:84861539512

SN - 9781467309615

SP - 400

EP - 403

BT - Proceedings - 2012 IEEE 8th International Colloquium on Signal Processing and Its Applications, CSPA 2012

ER -