Statistical Hyperspectral Detection: Signal Processing Perspectives

Comprehensive reference on hyperspectral target and anomaly detection discussing general theory and the latest HSI technological developments

Hyperspectral Target and Anomaly Detection provides detailed information on the evolution of general theory of hyperspectral target detection and anomaly detection from the past two decades, covering advanced HSI technologies that have been developed in hyperspectral data exploitation such as various new versions of OSP, CEM, and VD. This book pays special focus to statistical signal processing approaches in hyperspectral target and anomaly detection.

Hyperspectral Target and Anomaly Detection discusses topics including:

• Fundamental principles for hyperspectral imaging, including the hyperspectral binary communication channel and applications of the pigeon-hole and orthogonality principles
• Passive anomaly detection, covering endmember finding and target-to-anomaly deletion conversions
• Matrix decomposition models, including low-rank and sparse subspace decomposition, background-anomaly decomposition analysis, and rank estimation for model orders
• Pure-pixel, constrained energy minimization subpixel, and background-annihilated TCIMF target detection
• Statistical hyperspectral image classification, covering multiple hypothesis testing, CEM-based and LCMX-based classifiers, confusion matrices, and 3D ROC analysis

Hyperspectral Target and Anomaly Detection is a unique and up-to-date reference on the subject for students in electrical engineering and computer science as well as professionals and researchers in the fields of remote sensing, photogrammetry, geology, and forestry.