A Comparative Study of Descriptors and Detectors in Multispectral Face Recognition

Feature extraction is widely used in many computer vision problems such as object detection, image retrieval, object and scene recognition, etc. Feature extraction based on local feature detectors and descriptors requires a good compromise between the detectors and the descriptors. This paper presents a set of tools which are BRISK, FAST, FREAK, HARRIS and SURF for multispectral face recognition which is an alternative to conventional face recognition. The multispectral images are obtained by fusion of thermal and visible images.