Nowadays for biometric authentication system are used for security applications like verification and identification. Fingerprint, face and iris are the various biometric traits. Since the biometric traits are unique in nature, it is possible to avoid problems like password stolen or forgotten. Biometric has the capability to distinguish between real and fake. In this for software based fingerprint liveness detection we use Local Binary Pattern (LBP) for texture classification and Histogram of Oriented Gradient (HOG) for object detection and uses Support Vector Machine (SVM) classifies for classification. Through classification it is possible to distinguish between real and fake.

Now a days for security applications like identification and verification Biometric authentication systems are used. Biometric features are unique in nature, so this can be used to avoid typical problems of the system based on the use of password which can be stolen or forgotten. The research in this field is very active with local descriptors based on analysis of micro textural features in order to keep good level of security, gaining more and more popularity because of their excellent performance and flexibility. And this paper aims for accessing these descriptors for liveness detection in security systems based on various biometric traits: fingerprint, iris and face. In this work implements and evaluate two different feature extraction techniques for fingerprint and iris detection: -HOG and SURF descriptor respectively. Both the techniques were used in conjunction with a SVM classifier.