Local Grouped Order Pattern and Non-local Binary Pattern for Face Spoofing Detection

Face spoofing detection is an area of research which focuses on improving the security of facial recognition systems. It’s a cybersecurity strategy that prevents facial recognition systems from being fooled by fraudulent access. In this paper, a new face spoofing detection pipeline is proposed based on the extraction of Local Grouped Order Pattern and Non-local Binary Pattern (LGONBP) texture descriptors from the HSV color space channels. The LGONBP features are computed separately for each of the three channels in the HSV face image, the three generated texture descriptors are concatenated and passed into an SVM classifier. The publicly available MSU-MFSD database is used to evaluate the proposed method, the obtained results show competitive performance compared to the existing face spoofing detection approaches. computed independently for each of the three channels in the HSV face image.