Biometric Security System for Fake Detection using image Quality Assessment Techniques

Abstract

Biometric Identification is the best security system in developing security world. Hackers made a fake biometric in easy way, so it reduce the security accuracy. Biometric security system are hacked by using two types of attack, direct or spoofing attack & indirect attack. So, there is a need to develop a new and efficient protection measure. Here, three biometric techniques - face recognition, fingerprint and iris recognition (Multi-biometric system) are used to detect whether the image is real or fake. This protection method uses Image Quality Assessment (IQA) technique. If the image is a real, it checks whether the image belongs to an authorized person or not. Out of the three biometric techniques, face recognition using Discriminate Power Analysis (DPA) technique is used for detecting authorized person. The objective is to increase the security of biometric recognition framework by adding liveness assessment in fast, non-invasive & user friendly manner. This approach protects the biometric security system from direct and indirect attack. The complexity of the proposed system is very low because it extracts 31 image quality measures from one image.