Algorithms for Hyperspectral Processing of Transient Events

Period of Performance: 07/31/2006 - 11/30/2008

$744K

Phase 2 SBIR

Recipient Firm

Space Computer Corp.
12121 Wilshire Boulevard
Los Angeles, CA 90025
Principal Investigator

Abstract

Persistent hyperspectral imaging has important military applications including: 1) rapid detection and identification of transient battlefield events (e.g., weapon launches); 2) detection of low-observable moving targets not readily detected using conventional spatial or spectral discriminants, or that move too quickly to be acquired and followed by low revisit rate HSI sensors; and 3) dynamic tracking and identification of gaseous effluent plumes associated with chemical weapons. To realize the full potential of the emerging generation of staring video-HSI (VHSI) sensors, it is necessary to develop real-time calibration, correction, image cube formation and exploitation processing concepts that take full advantage of the novel data products these instruments provide. Space Computer Corporation and their subcontractor Bodkin Design & Engineering, LLC propose to develop a prototype solution to enable FPGA-based image cube formation and exploitation for VHSI sensors. The Phase I effort defined top-level VHSI operational issues, developed algorithm approaches, and demonstrated algorithm on VHSI data. In Phase II, we will implement the image calibration, correction and cube formation processing on an FPGA and integrate this with a personal computer-based exploitation prototype. This prototype will be used to demonstrate potential real-time VHSI capabilities.