Real Time Face Liveness Detection

Abstract: Face recognition is a widely used biometric approach. But face recognition systems are vulnerable to spoof attacks made by non-real faces, where a photo or video of an authorized person's face could be used to gain access to facilities or services. A secure system needs liveness detection in order to guard against such spoofing attacks. Here an efficient real time face liveness detection algorithm based on image distortion analysis (IDA) is proposed. Two different features such as blurriness and chromatic moment are extracted from the image. A fuzzy classifier is used to distinguish between live and spoof faces. The proposed approach is implemented in MATLAB 2013a tool box. Keywords: Face recognition, liveness detection, spoofing, image distortion analysis, fuzzy classifier.