Browse by topic
Type of publication
Illumination Invariance for Local Feature Face Recognition. 1st Spanish Workshop on Biometrics, 2007.Illumination invariance is one of the most difficult properties to achieve in a face recognition system. Illumination normalization is a way to solve this problem. Previous research has shown that local normalization methods are capable of reducing the error rates significantly even when there are extreme illumination changes. In this paper we propose an improvement to the local feature face recognition algorithm by previously doing an illumination normalization process. We show the results of some experiments carried out using the XM2VTS database and the Lausanne protocol for face verification.