No abstract

No abstract

No abstract

Human ear recognition is a new biometric technology which competes with other powerful biometrics modalities such as fingerprint, face and iris. In this paper we present a hypridised approach for ear biometric feature extraction named DWT-SIFT based on the combination of global and local approach named Wavelets and SIFT respectively. The proposed approach has been evaluated on two ear biometric databases, namely IIT Delhi and USTB 2. For performance evaluation of the proposed method we compute the false rejection rate (FRR), the false acceptance rate (FAR), accuracy and the needed time for ear authentication. Experimental results show that the proposed approach allows getting a higher accuracy and less time consumption compared to basic SIFT and wavelets based ear authentication systems taken individually.

No abstract