Face Recognition Across the Imaging Spectrum

This authoritative text/reference presents acomprehensive review of algorithms and techniques for face recognition (FR),with an emphasis on systems that can be reliably used in operationalenvironments. Insights are provided by an international team of pre-eminentexperts into the processing of multispectral and hyperspectral face imagescaptured under uncontrolled environments. These discussions cover a variety ofimaging sensors ranging from state-of-the-art visible and infrared imagingsensors, to RGB-D and mobile phone image sensors. A range of differentbiometric modalities are also examined, including face, periocular and iris.Topics andfeatures:Presents a detailed introduction to the spectralimaging of human skin, and describes the data collection of multispectral andhyperspectral face imagesCovers a range of face-based identificationtechnologies addressing challenges in the processing of face images in visibleand infrared spectral bandsProvides an overview of spoofing attacks and counter-measuresfor FR systems, and analyzes the effects of digital image alterations on FRalgorithmsInvestigates the processing of degraded facephotographs from ID documents, and examines mobile FR in still images and videosequencesProposes a state-of-the-art RGB-D FR algorithm, and suggestsalgorithms for 3D-based FR as well as techniques blending 2D and 3D dataDiscusses score normalization methods, and exploresthe use of patterns in the sclera in order to improve the robustness of irisrecognition systemsThis timely volume is a mine of useful information forresearchers, practitioners and students involved in image processing, computervision, biometrics and security.Dr. Thirimachos Bourlai is an AssistantProfessor in the Lane Department of Computer Science and Electrical Engineeringat West Virginia University, Morgantown, WV, USA.