Face spoofing detection using local binary patterns and Fisher Score

Azeddine Benlamoudi , Djamel Samai , Abdelkrim Ouafi , Salah Eddine Bekhouche , Abdelmalik Taleb-Ahmed (1) , Abdenour Hadid (2)

1 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201
2 CMV - Center for Machine Vision Research

Todays biometric systems are vulnerable to spoofattacks made by non-real faces. The problem is when a personshows in front of camera a print photo or a picture from cellphone. We study in this paper an anti-spoofing solution fordistinguishing between ’live’ and ’fake’ faces. In our approachwe used overlapping block LBP operator to extract features ineach region of the image. To reduce the features we used Fisher-Score. Finally, we used a nonlinear Support Vector Machine(SVM) classifier with kernel function for determining whether theinput image corresponds to a live face or not. Our experimentalanalysis on a publicly available NUAA and CASIA face anti-spoofing databases following the standard protocols showed goodresults

Mots clés: "biometric"; "spoofing"; "LBP"; "Fisher-Score"; "SVM"

Domaines: Automatique / Robotique

Azeddine Benlamoudi, Djamel Samai, Abdelkrim Ouafi, Salah Eddine Bekhouche, Abdelmalik Taleb-Ahmed, et al.. Face spoofing detection using local binary patterns and Fisher Score. 2015 3rd International Conference on Control, Engineering & Information Technology (CEIT), May 2015, Tlemcen, Algeria. pp.1-5, ⟨10.1109/CEIT.2015.7233145⟩. ⟨hal-03417090⟩