Detection Algorithms for Hyperspectral Imaging Applications

Abstract

Detection and identification of military and civilian targets from airborne platforms using hyperspectral sensors is of great interest. Relative to multispectral sensing, hyperspectral sensing can increase the detectability of pixel and subpixel size targets by exploiting finer detail in the spectral signatures of targets and natural backgrounds. A multitude of adaptive detection algorithms for resolved and subpixel targets, with known or unknown spectral characterization, in a background with known or unknown statistics, theoretically justified or ad hoc, with low or high computational complexity, have appeared in the literature or have found their way into software packages and end-user systems. The purpose of this report is twofold. First, we present a unified mathematical treatment of most adaptive matched filter detectors using common notation, and we state clearly the underlying theoretical assumptions. Whenever possible, we express existing ad hoc algorithms as computationally simpler versions of optimal methods. Second, we present a comparative performance analysis of the basic algorithms using theoretically obtained performance characteristics. We focus on algorithms characterized by theoretically desirable properties, practically desired features, or implementation simplicity. Sufficient detail is provided for others to verify and expand this evaluation and framework. A primary goal is to identify best-of-class algorithms for detailed performance evaluation. Finally, we provide a taxonomy of the key algorithms and introduce a preliminary experimental framework for evaluating their performance.

Optical Imaging and Measurement of Angiogenesis

“…The spectral data can be processed in a number of ways to accomplish different goals. Pixel classification is performed for assignment of pixels to distinct classes based on known pure reference spectra or spectra mathematically determined to be distinct based on the spectral content of the data set, while spectral unmixing is performed to determine the relative or absolute amount of a number of different spectrally distinct substances present at each pixel location using either a linear or a nonlinear combination of the spectra [94][95][96][97][98][99][100][101][102]. The processed spectral information can then be placed into a spatial context in an image.…”

A real-time CFAR thresholding method for target detection in hyperspectral images

“…By accounting for the theoretical statistical characteristics of detector's outputs, CFAR detector could obtain a CFAR threshold according to a pre-defined probability of false alarm (Pfa). Therefore, the accuracy of the outputs distribution modelling greatly affects the precision of the CFAR threshold determination [15,16].…”

Fake modern Chinese painting identification based on spectral–spatial feature fusion on hyperspectral image

“…This enables discrimination of materials based on the spectral characteristics (Ren et al 2014;Meijun et al 2015). HSI has been successfully applied in a wide range of application areas, such as geography (Cheng et al 2014Han et al 2014b;Zabalza et al 2015), military (Manolakis and Shaw 2002;Stein et al 2002;Eismann et al 2009), agriculture (Patel et al 2001;Datt et al 2003) and mineralogy (Hörig et al 2001).…”