Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)

These day, face recognition became main attraction by many researcher due to demand from commercial and law enforcement sectors. The main issues in face recognition are the sensitivity toward intrinsic factors and extrinsic factors. Beside, computation time and memory usage are the important aspect...

Full description

Main Authors: Selamat, M.H., Rais, H.M.
Format: Conference or Workshop Item
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-eprints.30512 /
Published: Institute of Electrical and Electronics Engineers Inc. 2016
Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85010366658&doi=10.1109%2fICCOINS.2016.7783253&partnerID=40&md5=179c196a9bfe7913339067f63418e5a8
http://eprints.utp.edu.my/30512/
Tags: Add Tag
No Tags, Be the first to tag this record!
id utp-eprints.30512
recordtype eprints
spelling utp-eprints.305122022-03-25T07:09:41Z Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM) Selamat, M.H. Rais, H.M. These day, face recognition became main attraction by many researcher due to demand from commercial and law enforcement sectors. The main issues in face recognition are the sensitivity toward intrinsic factors and extrinsic factors. Beside, computation time and memory usage are the important aspect to been consider. This research objective focus toward enhanced on Hybrid Multiclass Support Vector Machine (HM-SVM) model. The model consist of several phases which are data preparation, feature extraction and classification. In feature extraction phase, Discrete Wavelet Transform (DWT) and Principle Component Analysis (PCA) was applied to perform image compression, dimension reduction and feature extraction. Then, Hybrid Multiclass Support Vector Machine (HM-SVM) strategies were utilized to tackle the face recognition problem. Cambridge ORL Face Database was used as testing and training material. This database consist of 400 images of 40 individuals. The accuracy evaluation of this research was based on two different SVM kernel. Besides, the research was performed based on Orthogonal Wavelet families. Comparison was made with classic Hybrid Multiclass Support Vector Machine (HM-SVM) performance. As a result, the proposed algorithm shown the enhancement improve the classic HS-SVM approximately by 3.13-6.56 using polynomial kernel and 1.88-5.31 by using radial basis function kernel (RBF). © 2016 IEEE. Institute of Electrical and Electronics Engineers Inc. 2016 Conference or Workshop Item NonPeerReviewed https://www.scopus.com/inward/record.uri?eid=2-s2.0-85010366658&doi=10.1109%2fICCOINS.2016.7783253&partnerID=40&md5=179c196a9bfe7913339067f63418e5a8 Selamat, M.H. and Rais, H.M. (2016) Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM). In: UNSPECIFIED. http://eprints.utp.edu.my/30512/
institution Universiti Teknologi Petronas
collection UTP Institutional Repository
description These day, face recognition became main attraction by many researcher due to demand from commercial and law enforcement sectors. The main issues in face recognition are the sensitivity toward intrinsic factors and extrinsic factors. Beside, computation time and memory usage are the important aspect to been consider. This research objective focus toward enhanced on Hybrid Multiclass Support Vector Machine (HM-SVM) model. The model consist of several phases which are data preparation, feature extraction and classification. In feature extraction phase, Discrete Wavelet Transform (DWT) and Principle Component Analysis (PCA) was applied to perform image compression, dimension reduction and feature extraction. Then, Hybrid Multiclass Support Vector Machine (HM-SVM) strategies were utilized to tackle the face recognition problem. Cambridge ORL Face Database was used as testing and training material. This database consist of 400 images of 40 individuals. The accuracy evaluation of this research was based on two different SVM kernel. Besides, the research was performed based on Orthogonal Wavelet families. Comparison was made with classic Hybrid Multiclass Support Vector Machine (HM-SVM) performance. As a result, the proposed algorithm shown the enhancement improve the classic HS-SVM approximately by 3.13-6.56 using polynomial kernel and 1.88-5.31 by using radial basis function kernel (RBF). © 2016 IEEE.
format Conference or Workshop Item
author Selamat, M.H.
Rais, H.M.
spellingShingle Selamat, M.H.
Rais, H.M.
Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)
author_sort Selamat, M.H.
title Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)
title_short Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)
title_full Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)
title_fullStr Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)
title_full_unstemmed Enhancement on image face recognition using Hybrid Multiclass SVM (HM-SVM)
title_sort enhancement on image face recognition using hybrid multiclass svm (hm-svm)
publisher Institute of Electrical and Electronics Engineers Inc.
publishDate 2016
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85010366658&doi=10.1109%2fICCOINS.2016.7783253&partnerID=40&md5=179c196a9bfe7913339067f63418e5a8
http://eprints.utp.edu.my/30512/
_version_ 1741197419430805504
score 11.62408

Similar Items