Wavelength bands selection method of multispectral target detection

A framework was proposed for the selection of wavelength bands for multispectral sensors by use of hyperspectral reference data. Using the results from the detection theory, a cost function was derived that was minimized by a set of spectral bands optimal in terms of detection performance for discrimination between a class of small rare targets and clutter with known spectral distribution. An approximate minimizer of cost function is found by using matching pursuit algorithm, and the filters are selected. The method may be used in the design of multispectral infrared search and track and electro-optical missile warning sensors, where a low false-alarm rate and a high-detection probability for detection of small targets against a clutter background are of critical importance.