Spectral face recognition using orthogonal subspace bases

Andrew Wimberly, Stefan Robila, Tansy Peplau

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

3 Citations (Scopus)

Abstract

We present an efficient method for facial recognition using hyperspectral imaging and orthogonal subspaces. Projecting the data into orthogonal subspaces has the advantage of compactness and reduction of redundancy. We focus on two approaches: Principal Component Analysis and Orthogonal Subspace Projection. Our work is separated in three stages. First, we designed an experimental setup that allowed us to create a hyperspectral image database of 17 subjects under different facial expressions and viewing angles. Second, we investigated approaches to employ spectral information for the generation of fused grayscale images. Third, we designed and tested a recognition system based on the methods described above. The experimental results show that spectral fusion leads to improvement of recognition accuracy when compared to regular imaging. The work expands on previous band extraction research and has the distinct advantage of being one of the first that combines spatial information (i.e. face characteristics) with spectral information. In addition, the techniques are general enough to accommodate differences in skin spectra.

Original languageEnglish
Title of host publicationAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI
Volume7695
DOIs
StatePublished - 25 Jun 2010
EventAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI - Orlando, FL, United States
Duration: 5 Apr 20108 Apr 2010

Other

OtherAlgorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI
CountryUnited States
CityOrlando, FL
Period5/04/108/04/10

Fingerprint

Face recognition
Face Recognition
Principal component analysis
Redundancy
Skin
Fusion reactions
Subspace
Imaging techniques
Hyperspectral Imaging
Hyperspectral Image
Facial Expression
Image Database
Spatial Information
void ratio
redundancy
principal components analysis
Principal Component Analysis
Expand
Compactness
Fusion

Keywords

  • Face recognition
  • Hyperspectral data
  • Orthogonal subspaces
  • Principal Component Analysis

Cite this

Wimberly, A., Robila, S., & Peplau, T. (2010). Spectral face recognition using orthogonal subspace bases. In Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI (Vol. 7695). [76952E] https://doi.org/10.1117/12.849892
Wimberly, Andrew ; Robila, Stefan ; Peplau, Tansy. / Spectral face recognition using orthogonal subspace bases. Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI. Vol. 7695 2010.
@inproceedings{98393728bee946138b5863528b3d1cc6,
title = "Spectral face recognition using orthogonal subspace bases",
abstract = "We present an efficient method for facial recognition using hyperspectral imaging and orthogonal subspaces. Projecting the data into orthogonal subspaces has the advantage of compactness and reduction of redundancy. We focus on two approaches: Principal Component Analysis and Orthogonal Subspace Projection. Our work is separated in three stages. First, we designed an experimental setup that allowed us to create a hyperspectral image database of 17 subjects under different facial expressions and viewing angles. Second, we investigated approaches to employ spectral information for the generation of fused grayscale images. Third, we designed and tested a recognition system based on the methods described above. The experimental results show that spectral fusion leads to improvement of recognition accuracy when compared to regular imaging. The work expands on previous band extraction research and has the distinct advantage of being one of the first that combines spatial information (i.e. face characteristics) with spectral information. In addition, the techniques are general enough to accommodate differences in skin spectra.",
keywords = "Face recognition, Hyperspectral data, Orthogonal subspaces, Principal Component Analysis",
author = "Andrew Wimberly and Stefan Robila and Tansy Peplau",
year = "2010",
month = "6",
day = "25",
doi = "10.1117/12.849892",
language = "English",
isbn = "9780819481597",
volume = "7695",
booktitle = "Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI",

}

Wimberly, A, Robila, S & Peplau, T 2010, Spectral face recognition using orthogonal subspace bases. in Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI. vol. 7695, 76952E, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI, Orlando, FL, United States, 5/04/10. https://doi.org/10.1117/12.849892

Spectral face recognition using orthogonal subspace bases. / Wimberly, Andrew; Robila, Stefan; Peplau, Tansy.

Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI. Vol. 7695 2010. 76952E.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Spectral face recognition using orthogonal subspace bases

AU - Wimberly, Andrew

AU - Robila, Stefan

AU - Peplau, Tansy

PY - 2010/6/25

Y1 - 2010/6/25

N2 - We present an efficient method for facial recognition using hyperspectral imaging and orthogonal subspaces. Projecting the data into orthogonal subspaces has the advantage of compactness and reduction of redundancy. We focus on two approaches: Principal Component Analysis and Orthogonal Subspace Projection. Our work is separated in three stages. First, we designed an experimental setup that allowed us to create a hyperspectral image database of 17 subjects under different facial expressions and viewing angles. Second, we investigated approaches to employ spectral information for the generation of fused grayscale images. Third, we designed and tested a recognition system based on the methods described above. The experimental results show that spectral fusion leads to improvement of recognition accuracy when compared to regular imaging. The work expands on previous band extraction research and has the distinct advantage of being one of the first that combines spatial information (i.e. face characteristics) with spectral information. In addition, the techniques are general enough to accommodate differences in skin spectra.

AB - We present an efficient method for facial recognition using hyperspectral imaging and orthogonal subspaces. Projecting the data into orthogonal subspaces has the advantage of compactness and reduction of redundancy. We focus on two approaches: Principal Component Analysis and Orthogonal Subspace Projection. Our work is separated in three stages. First, we designed an experimental setup that allowed us to create a hyperspectral image database of 17 subjects under different facial expressions and viewing angles. Second, we investigated approaches to employ spectral information for the generation of fused grayscale images. Third, we designed and tested a recognition system based on the methods described above. The experimental results show that spectral fusion leads to improvement of recognition accuracy when compared to regular imaging. The work expands on previous band extraction research and has the distinct advantage of being one of the first that combines spatial information (i.e. face characteristics) with spectral information. In addition, the techniques are general enough to accommodate differences in skin spectra.

KW - Face recognition

KW - Hyperspectral data

KW - Orthogonal subspaces

KW - Principal Component Analysis

UR - http://www.scopus.com/inward/record.url?scp=77953760094&partnerID=8YFLogxK

U2 - 10.1117/12.849892

DO - 10.1117/12.849892

M3 - Conference contribution

SN - 9780819481597

VL - 7695

BT - Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI

ER -

Wimberly A, Robila S, Peplau T. Spectral face recognition using orthogonal subspace bases. In Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI. Vol. 7695. 2010. 76952E https://doi.org/10.1117/12.849892