z-logo
open-access-imgOpen Access
Face Liveness Detection using Feature Fusion Using Block Truncation Code Technique
Author(s) -
Prasad A. Jagdale,
Sudeep D. Thepade
Publication year - 2019
Publication title -
international journal on recent and innovation trends in computing and communication
Language(s) - English
Resource type - Journals
ISSN - 2321-8169
DOI - 10.17762/ijritcc.v7i8.5348
Subject(s) - spoofing attack , computer science , liveness , biometrics , facial recognition system , password , artificial intelligence , face (sociological concept) , feature (linguistics) , block (permutation group theory) , three dimensional face recognition , computer vision , pattern recognition (psychology) , adaboost , face detection , computer security , support vector machine , social science , linguistics , philosophy , geometry , mathematics , sociology , programming language
Nowadays the system which holds private and confidential data are being protected using biometric password such as finger recognition, voice recognition, eyries and face recognition. Face recognition match the current user face with faces present in the database of that security system and it has one major drawback that it never works better if it doesn’t have liveness detection. These face recognition system can be spoofed using various traits. Spoofing is accessing a system software or data by harming the biometric recognition security system. These biometric systems can be easily attacked by spoofs like peoples face images, masks and videos which are easily available from social media. The proposed work mainly focused on detecting the spoofing attack by training the system. Spoofing methods like photo, mask or video image can be easily identified by this method. This paper proposed a fusion technique where different features of an image are combining together so that it can give best accuracy in terms of distinguish between spoof and live face. Also a comparative study is done of machine learning classifiers to find out which classifiers gives best accuracy.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here