Face Presentation Attack Detection Based On Distortion Analysis

Face recognition(FR)technologies have been widely used in identity authentication.However,with the development of information technologies,images or videos containing a target's face can be easily acquired from online social network nowadays.If the facial disclosures misused by malicious attackers,it is possible to launch presentation attack(also known as spoofing attack)to FR system.To counter photo attack and video attack in FR system,this work analyzed the processing of common face presentation attack and its influence on the face image.The distortion analysis method in face presentation detection(PAD)is elaborated.Based on the distortion difference between real face and artefact,photo attack and video attack can be effectively detected.To sum up,the contributions of this work are:1)In order to minimize the influence of the redundant noise contamination,guided scale space is proposed to reduce the redundancy of the original facial texture and to extract more powerful facial edges.Based on the guided scale space,two guided scale texture descriptors are proposed to extract liveness detection features,and they are guided scale based local binary pattern and local guided binary pattern.With the guided scale texture features,a face PAD sheme with guided scale texture is proposed.The experimental results indicate the effectiveness of the proposed method for countering photo attack and video attack in FR.2)In order to fully analyze the chromatic facial texture,the chromatic co-occurrence of local binary pattern(CCoLBP)is proposed to investigate the inter-channel based information.With the fusion of intra-channel based facial texture and the CCoLBP feature,a face PAD sheme based on color degradation analysis is proposed,and the differences of color degradation and texture distribution between real face and artefact can be characterized.The experimental results demonstrate the promising performance of the proposed method,especially in cross-database testing.3)A face PAD protosystem is implemented.The proposed method is fast,user friendly,and has the potential to be applied for improving the security of FR system.