Logic Fusion of Color based on new Fast Feature extraction for face authentication

Identity verification using face information is a challenging research area that was very active recently, mainly because of its natural and non intrusive interaction with the authentication system. Principal Component Analysis (PCA) is a typical face based method which considers face as global feature. In this paper, we propose a new face based authentication approach based on the use of face image mean and standard deviation (MSD) as feature vector. Once the feature vector is extracted, the next stage consists in comparing it with the feature vector of the claimed client face. To demonstrate the efficiency of the MSD approach many experiments have been done using XM2VTS database according to the protocol of Lausanne. The obtained results show that the proposed method is more efficient and faster than PCA.