Automatic target detection and recognition in multiband imagery: a unified ML detection and estimation approach

Multispectral or hyperspectral sensors can facilitate automatic target detection and recognition (ATD/R) in clutter since the natural clutter from vegetation is characterized by a grey body, and man-made objects, compared with blackbody radiators, emit radiation more strongly at some wavelengths than at others. Various types of data fusion of the spectral-spatial features contained in multiband imagery were developed during the last decade for detecting and recognizing low-contrast targets in a clutter background. While different approaches to detection were taken for a variety of problem scenarios, they appear to have a common framework. In this paper, a generalized hypothesis test on the observed data is formulated by partitioning the received bands into two groups. In one group, targets exhibit substantial coloring in their signatures but behave either like grey bodies or emit negligible radiant energy in the other group. This general observation about the data generalizes the data models used previously. A unified framework for these problems, which utilizes a maximum likelihood ratio approach to detection, is presented herein.Within this framework, a performance evaluation and a comparison of the various types of multiband detectors are conducted by finding the gain of the signal-noise-ratio (SNR) needed for detection as well as the gain required for separability between the target classes used for recognition. Certain multiband detectors developed previously become special cases in this new, more general framework when the assumptions are simplified. The incremental gains in SNR and separability obtained by using what are called target-feature bands plus clutter-reference bands are studied. In addition, certain essential parameters are defined that effect the gains in SNR and target separability. Instead of incorporating a priori clutter statistics into a detector, this general framework for maximum likelihood ratio detection is able to adapt automatically to the local clutter statistics.