Abstract
We demonstrate the applicability of integrated sensing and processing decision trees (ISPDTs) methodology to a set of digital mirror array (DMA) hyperspectral imagery. In particular, we demonstrate that ISPDTs can be used to detect and localize targets by using just a few DMA Hadamard frames, so that an entire hyperspectral data cube need not be collected to successfully perform the given task. This suggests that such an integrated sensing–processing suite may be appropriate for extremely time-sensitive pattern-recognition applications.
© 2006 Optical Society of America
Full Article | PDF ArticleMore Like This
Bernard R. Foy, James Theiler, and Andrew M. Fraser
Opt. Express 17(20) 17391-17411 (2009)
Luis Guanter, Rudolf Richter, and José Moreno
Appl. Opt. 45(10) 2360-2370 (2006)
D. D. Saxena and J. R. Saylor
Appl. Opt. 45(12) 2672-2688 (2006)