Faculté des sciences
Texte intégral

Comparison of feature extraction techniques for face verification using elastic graph matching on low-power mobile devices

Nagel, Jean-Luc ; Stadelmann, Patrick ; Ansorge, Michael ; Pellandini, Fausto

In: EUROCON 2003 (Computer as a Tool. The IEEE Region 8), 2003, vol. 2, p. 365-369

Biometric face verification using elastic graph matching is considered and two feature extraction techniques used as a pre-processing stage are compared. Low-complexity vs. performance is of particular interest for mobile devices. A modified technique based on a normalization of mathematical morphology feature is introduced and show a performance comparable to that of Gabor features at a... More

Add to personal list
    Title
    • Comparison of feature extraction techniques for face verification using elastic graph matching on low-power mobile devices
    Author
    • Nagel, Jean-Luc. Institut de Microtechnique, Université de Neuchâtel, Switzerland
    • Stadelmann, Patrick. Institut de Microtechnique, Université de Neuchâtel, Switzerland
    • Ansorge, Michael. Institut de Microtechnique, Université de Neuchâtel, Switzerland
    • Pellandini, Fausto. Institut de Microtechnique, Université de Neuchâtel, Switzerland
    Document Type
    • Postprint
    Publication Date
    • 2003-09-22
    Language
    • English
    Published in
    • EUROCON 2003 (Computer as a Tool. The IEEE Region 8), 2003, vol. 2, p. 365-369. Institute of Electrical and Electronics Engineers (IEEE)
    Other Version
    OAI-PMH Identifier
    • oai:doc.rero.ch:20061221150052-OI
    Submission No.
    • 20061221150052-OI
    Summary
    Biometric face verification using elastic graph matching is considered and two feature extraction techniques used as a pre-processing stage are compared. Low-complexity vs. performance is of particular interest for mobile devices. A modified technique based on a normalization of mathematical morphology feature is introduced and show a performance comparable to that of Gabor features at a much-reduced computational complexity.