Multifeature Biometric Based Authentication System

Abstract Authentication a system by biometrics is one of the systematic way of identifying a person without any spoofing. Multifeature biometric palm vein authentication is in the progression of being endlessly refined and developed. Multispectral palm vein biometric system has its applications over many areas which includes the access control, hospitals, bank and also in application of forensic. An individuals palm vein image is converted into data points by algorithms, which is then compressed, encrypted, and stored by the software. Whenever a person access it, the data is compared with the stored one and thus verification is done. This work proposes a multi feature palm vein recognition method where the features of the input images are extracted with the efficient methodologies like Spatio local binary pattern, which is the advanced technology of LPB and modified radon transform. Once after the features are extracted they are classified. This classification is done using Support Vector Machine (SVM) algorithm. The proposed method effectively accommodates the rotational, potential deformations and translational changes by encoding the orientation conserving features. The proposed system examines multifeature biometric based authentication system using a setup where the images are said to be contactless. These databases were obtained from CASIA and IIT Delhi. The results obtained are said to be effecient than any other results obtained using different methodologies in earlier papers discussed in the literature survey.