Fusion of visible and infrared image features for face recognition

Abstract

In this paper we used Local Binary Pattern (LBP), Features from accelerated segment test (FAST), Scale Invariant Features Transformations (SIFT), Speed Up Robust Feature Transformations (SURF), Binary Robust Invariant Scalable Key points (BRISK), Maximally Stable Extremal Regions (MSER) feature extraction methods to evaluate the performance of face recognition system with fusion of visible and infrared images. These six feature extraction methods are tested and analyzed on OTCBVS database under various illumination and expressions of multiple persons. The results shows that FAST, SIFT and SURF provides high precision rate and recall rate than other methods.