Evaluation of Skin Spectral Features for Biometric
| Affiliation auteurs | !!!! Error affiliation !!!! |
| Titre | Evaluation of Skin Spectral Features for Biometric |
| Type de publication | Conference Paper |
| Year of Publication | 2017 |
| Auteurs | Li C, Benezeth Y, Nakamura K, Gomez R, Yang F |
| Conference Name | 2017 IEEE 2ND INTERNATIONAL CONFERENCE ON SIGNAL AND IMAGE PROCESSING (ICSIP) |
| Publisher | IEEE |
| Conference Location | 345 E 47TH ST, NEW YORK, NY 10017 USA |
| ISBN Number | 978-1-5386-0969-9 |
| Mots-clés | biometric authentication, Mahalanobis distance, Multispectral image, skin spectrum classification |
| Résumé | In the recent years, multispectral imaging has been successfully used in various biometric authentication applications. However, in most cases, the frames of multispectral images are consolidated simply by using data fusion techniques rather than contributing directly to the recognition process. This paper evaluates the viability of face skin spectrum for biometric authentication by comparing inter-and intra-user spectral distance. The experiment of this work is conducted using the hyperspectral face database of the Stanford Center for Image Systems Engineering (SCIEN). Obtained results demonstrate that, even if not sufficient by itself, human skin spectra contain useful information to the person authentications, and it potentially allows to improve the accuracy and ergonomics performances of biometric systems. |