Prevention of Spoof Attack in Biometric System Using Liveness Detection
Sanjeevakumar M. Hatture, Nalinakshi B. G, Rashmi P. Karchi · 2013
Biometric system has gained wide range of motivations and applications in security domain. Biometric systems relay on the biometric characteristics/data taken from the user for authentication. Unfortunately such biometric data is stolen or duplicated by the imposters/unauthorized users. Most of the biometrics systems depend purely on identifying the physiological characteristics of the user. It becomes easier to spoof in these biometric systems with the aid of fake biometric; it further reduces the reliability and security of biometric system. Spoof fools the system through the process of deception and impersonating others to make out that they are authorized in order to gain access in to the biometric system. Now a day's spoofing has become very common on the internet which thus leads to identify theft and fraud. There are several level of spoofing attacks such as placing fake biometrics on the sensor, replay attack, Attacking the enrolment centre Corrupting the matcher, Attacking the application etc, these in turn will reduce the level of security and reliability of biometric system. Liveness identification using the facial features also has been receiving more attention compared to other biometric modalities. Prevention of spoof attack in biometric system can be done by detecting the liveness of the user with the help of local facial features like eye blinking, lip movement, forehead and chin movement pattern of the face detected with real-time generic web-camera. In the proposed work, an effective authentication system using face biometric modality by developing the liveness detection model using the variations in the facial movements. The developed method is evaluated for 50 users, with severe variation in pose and expressions. With own database face recognition accuracy is around 86% and Liveness detection performance for the 50 authenticated users is around 88% is achieved.