The analysis for Gait Energy Image based on statistical methods

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

3 Citations (Scopus)

Abstract

Gait features representation is an important step in gait biometric recognition. It represents content of the individual image and the pattern of human walking simultaneously. Discriminative features can provide the most relevant and useful information for individual classification. Many state of the art feature fusions based on average silhouette were used for gait feature representation technique. Those fusion methods are mostly robust in simple case but inadequate for difficult gait representation such as walking with occlusion, walking with different view and walking with briefcase that affect the accuracy of gait performance. For that reason, we are exploring and proposing another potential feature fusion based on statistical method namely edges directed matrix of Gait Energy Image (EDMSGEI). We compare our EDMSGEI method with Gray Level Co-Occurrence Matrix (GLCMGEI) and Histogram of Gaussian (HOGGEI) method for individual recognition with K-Nearest neighbour classifier. The datasets involved are CASIA and UKM-CSM dataset images. Apart from HOGGEI, this study also shows that EDMSGEI gained better gait recognition performance.

Original languageEnglish
Title of host publication2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages125-128
Number of pages4
ISBN (Electronic)9781509028894
DOIs
Publication statusPublished - 27 Mar 2017
Event2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016 - Putrajaya, Malaysia
Duration: 14 Nov 201616 Nov 2016

Other

Other2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016
CountryMalaysia
CityPutrajaya
Period14/11/1616/11/16

Fingerprint

gait
Statistical methods
Fusion reactions
walking
Biometrics
fusion
energy
Classifiers
command service modules
biometrics
occlusion
matrices
classifiers
histograms
occurrences

Keywords

  • Gait Energy Image
  • Gait Recognition
  • Statistical method

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Biomedical Engineering
  • Control and Systems Engineering
  • Hardware and Architecture
  • Computer Networks and Communications
  • Instrumentation

Cite this

Rahman, S. Z. A., Sheikh Abdullah, S. N. H., & Ahmad Nazri, M. Z. (2017). The analysis for Gait Energy Image based on statistical methods. In 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016 (pp. 125-128). [7888022] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICAEES.2016.7888022

The analysis for Gait Energy Image based on statistical methods. / Rahman, Siti Zaharah Abd; Sheikh Abdullah, Siti Norul Huda; Ahmad Nazri, Mohd Zakree.

2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016. Institute of Electrical and Electronics Engineers Inc., 2017. p. 125-128 7888022.

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

Rahman, SZA, Sheikh Abdullah, SNH & Ahmad Nazri, MZ 2017, The analysis for Gait Energy Image based on statistical methods. in 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016., 7888022, Institute of Electrical and Electronics Engineers Inc., pp. 125-128, 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016, Putrajaya, Malaysia, 14/11/16. https://doi.org/10.1109/ICAEES.2016.7888022
Rahman SZA, Sheikh Abdullah SNH, Ahmad Nazri MZ. The analysis for Gait Energy Image based on statistical methods. In 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016. Institute of Electrical and Electronics Engineers Inc. 2017. p. 125-128. 7888022 https://doi.org/10.1109/ICAEES.2016.7888022
Rahman, Siti Zaharah Abd ; Sheikh Abdullah, Siti Norul Huda ; Ahmad Nazri, Mohd Zakree. / The analysis for Gait Energy Image based on statistical methods. 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 125-128
@inproceedings{5caa2e9f56f84531960b68edceaf9bcc,
title = "The analysis for Gait Energy Image based on statistical methods",
abstract = "Gait features representation is an important step in gait biometric recognition. It represents content of the individual image and the pattern of human walking simultaneously. Discriminative features can provide the most relevant and useful information for individual classification. Many state of the art feature fusions based on average silhouette were used for gait feature representation technique. Those fusion methods are mostly robust in simple case but inadequate for difficult gait representation such as walking with occlusion, walking with different view and walking with briefcase that affect the accuracy of gait performance. For that reason, we are exploring and proposing another potential feature fusion based on statistical method namely edges directed matrix of Gait Energy Image (EDMSGEI). We compare our EDMSGEI method with Gray Level Co-Occurrence Matrix (GLCMGEI) and Histogram of Gaussian (HOGGEI) method for individual recognition with K-Nearest neighbour classifier. The datasets involved are CASIA and UKM-CSM dataset images. Apart from HOGGEI, this study also shows that EDMSGEI gained better gait recognition performance.",
keywords = "Gait Energy Image, Gait Recognition, Statistical method",
author = "Rahman, {Siti Zaharah Abd} and {Sheikh Abdullah}, {Siti Norul Huda} and {Ahmad Nazri}, {Mohd Zakree}",
year = "2017",
month = "3",
day = "27",
doi = "10.1109/ICAEES.2016.7888022",
language = "English",
pages = "125--128",
booktitle = "2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - The analysis for Gait Energy Image based on statistical methods

AU - Rahman, Siti Zaharah Abd

AU - Sheikh Abdullah, Siti Norul Huda

AU - Ahmad Nazri, Mohd Zakree

PY - 2017/3/27

Y1 - 2017/3/27

N2 - Gait features representation is an important step in gait biometric recognition. It represents content of the individual image and the pattern of human walking simultaneously. Discriminative features can provide the most relevant and useful information for individual classification. Many state of the art feature fusions based on average silhouette were used for gait feature representation technique. Those fusion methods are mostly robust in simple case but inadequate for difficult gait representation such as walking with occlusion, walking with different view and walking with briefcase that affect the accuracy of gait performance. For that reason, we are exploring and proposing another potential feature fusion based on statistical method namely edges directed matrix of Gait Energy Image (EDMSGEI). We compare our EDMSGEI method with Gray Level Co-Occurrence Matrix (GLCMGEI) and Histogram of Gaussian (HOGGEI) method for individual recognition with K-Nearest neighbour classifier. The datasets involved are CASIA and UKM-CSM dataset images. Apart from HOGGEI, this study also shows that EDMSGEI gained better gait recognition performance.

AB - Gait features representation is an important step in gait biometric recognition. It represents content of the individual image and the pattern of human walking simultaneously. Discriminative features can provide the most relevant and useful information for individual classification. Many state of the art feature fusions based on average silhouette were used for gait feature representation technique. Those fusion methods are mostly robust in simple case but inadequate for difficult gait representation such as walking with occlusion, walking with different view and walking with briefcase that affect the accuracy of gait performance. For that reason, we are exploring and proposing another potential feature fusion based on statistical method namely edges directed matrix of Gait Energy Image (EDMSGEI). We compare our EDMSGEI method with Gray Level Co-Occurrence Matrix (GLCMGEI) and Histogram of Gaussian (HOGGEI) method for individual recognition with K-Nearest neighbour classifier. The datasets involved are CASIA and UKM-CSM dataset images. Apart from HOGGEI, this study also shows that EDMSGEI gained better gait recognition performance.

KW - Gait Energy Image

KW - Gait Recognition

KW - Statistical method

UR - http://www.scopus.com/inward/record.url?scp=85018170883&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018170883&partnerID=8YFLogxK

U2 - 10.1109/ICAEES.2016.7888022

DO - 10.1109/ICAEES.2016.7888022

M3 - Conference contribution

AN - SCOPUS:85018170883

SP - 125

EP - 128

BT - 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering, ICAEES 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -