A Novel Feature Descriptor for Face Anti-Spoofing Using Texture Based Method

Abstract

In this paper we propose a novel approach for face anti-spoofing called Extended Division Directional Ternary Co-relation Pattern (EDDTCP). The EDDTCP encodes co-relation of ternary edges based on the centre pixel gray values with its immediate directional neighbour and its next immediate average directional neighbour, which is calculated by using the average of cornered neighbours with directional neighbours. The proposed method is robust against presentation attacks by extracting the spatial information in all directions. Three Experiments were performed by using all the four texture descriptors (LBP, LTP, LGS and EDDTCP) and the results are compared. The proposed face anti-spoofing method performs better than LBP, LTP and LGS.