Quo vadis face recognition: Spectral considerations

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

1 Citation (Scopus)

Abstract

The paper provides novel approaches for the employment of spectral information when pursuing face recognition. We designed and tested Eigenface based algorithms that improve face recognition through feature extraction, i.e. extracting the 'best bands' according to various criteria such as decorelation and statistical independence. Eigenfaces correspond to principal components and have previously been used for regular grayscale and color images. In this paper we expand theiruse to hyperspectral imagery, i.e. data sets of images of the same scene associated to narrow wavelength intervals. Our approach is a two decomposition process. In the first, the hyperspectral data is reduced to grayscale using Principal Component Analysis. In the second, the grayscale images are processed using the classical Eigenface detection algorithm. The results suggest that spectral imaging improves face classification over its counterpart.

Original languageEnglish
Title of host publication2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009
DOIs
StatePublished - 25 Sep 2009
Event2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009 - Farmingdale, NY, United States
Duration: 1 May 20091 May 2009

Publication series

Name2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009

Other

Other2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009
CountryUnited States
CityFarmingdale, NY
Period1/05/091/05/09

Fingerprint

Face recognition
Principal component analysis
Feature extraction
Color
Decomposition
Imaging techniques
Wavelength

Keywords

  • Hyperspectral imaging
  • Image classification
  • Principal component analysis
  • Spectral analysis

Cite this

Robila, S. (2009). Quo vadis face recognition: Spectral considerations. In 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009 [5031555] (2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009). https://doi.org/10.1109/LISAT.2009.5031555
Robila, Stefan. / Quo vadis face recognition : Spectral considerations. 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009. 2009. (2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009).
@inproceedings{ee53aa154cb0459b8597587d67fe171c,
title = "Quo vadis face recognition: Spectral considerations",
abstract = "The paper provides novel approaches for the employment of spectral information when pursuing face recognition. We designed and tested Eigenface based algorithms that improve face recognition through feature extraction, i.e. extracting the 'best bands' according to various criteria such as decorelation and statistical independence. Eigenfaces correspond to principal components and have previously been used for regular grayscale and color images. In this paper we expand theiruse to hyperspectral imagery, i.e. data sets of images of the same scene associated to narrow wavelength intervals. Our approach is a two decomposition process. In the first, the hyperspectral data is reduced to grayscale using Principal Component Analysis. In the second, the grayscale images are processed using the classical Eigenface detection algorithm. The results suggest that spectral imaging improves face classification over its counterpart.",
keywords = "Hyperspectral imaging, Image classification, Principal component analysis, Spectral analysis",
author = "Stefan Robila",
year = "2009",
month = "9",
day = "25",
doi = "10.1109/LISAT.2009.5031555",
language = "English",
isbn = "9781424423484",
series = "2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009",
booktitle = "2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009",

}

Robila, S 2009, Quo vadis face recognition: Spectral considerations. in 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009., 5031555, 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009, 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009, Farmingdale, NY, United States, 1/05/09. https://doi.org/10.1109/LISAT.2009.5031555

Quo vadis face recognition : Spectral considerations. / Robila, Stefan.

2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009. 2009. 5031555 (2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009).

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

TY - GEN

T1 - Quo vadis face recognition

T2 - Spectral considerations

AU - Robila, Stefan

PY - 2009/9/25

Y1 - 2009/9/25

N2 - The paper provides novel approaches for the employment of spectral information when pursuing face recognition. We designed and tested Eigenface based algorithms that improve face recognition through feature extraction, i.e. extracting the 'best bands' according to various criteria such as decorelation and statistical independence. Eigenfaces correspond to principal components and have previously been used for regular grayscale and color images. In this paper we expand theiruse to hyperspectral imagery, i.e. data sets of images of the same scene associated to narrow wavelength intervals. Our approach is a two decomposition process. In the first, the hyperspectral data is reduced to grayscale using Principal Component Analysis. In the second, the grayscale images are processed using the classical Eigenface detection algorithm. The results suggest that spectral imaging improves face classification over its counterpart.

AB - The paper provides novel approaches for the employment of spectral information when pursuing face recognition. We designed and tested Eigenface based algorithms that improve face recognition through feature extraction, i.e. extracting the 'best bands' according to various criteria such as decorelation and statistical independence. Eigenfaces correspond to principal components and have previously been used for regular grayscale and color images. In this paper we expand theiruse to hyperspectral imagery, i.e. data sets of images of the same scene associated to narrow wavelength intervals. Our approach is a two decomposition process. In the first, the hyperspectral data is reduced to grayscale using Principal Component Analysis. In the second, the grayscale images are processed using the classical Eigenface detection algorithm. The results suggest that spectral imaging improves face classification over its counterpart.

KW - Hyperspectral imaging

KW - Image classification

KW - Principal component analysis

KW - Spectral analysis

UR - http://www.scopus.com/inward/record.url?scp=70349272750&partnerID=8YFLogxK

U2 - 10.1109/LISAT.2009.5031555

DO - 10.1109/LISAT.2009.5031555

M3 - Conference contribution

AN - SCOPUS:70349272750

SN - 9781424423484

T3 - 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009

BT - 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009

ER -

Robila S. Quo vadis face recognition: Spectral considerations. In 2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009. 2009. 5031555. (2009 IEEE Long Island Systems, Applications and Technology Conference, LISAT 2009). https://doi.org/10.1109/LISAT.2009.5031555