Spectral Calibration, Spatial Mapping and Flat Fielding Studies of the Dark HORSE 1 (DH1) March Data Collection

Abstract

Abstract : Recently researchers at the Naval Research Laboratory (NRL) have successfully demonstrated autonomous, real time, hyperspectral detection of airborne and military ground targets. Real-time autonomous cueing of a high-resolution imager, and designation of targets with pointing optics and a pulsed aser was also demonstrated. The work was performed under NRL's Dark HORSE effort, the culmination of a four-year Multispectral Overhead IR/EO Surveillance (MOVIES) program to develop real-time hyperspectral detection, cueing, target location, and target designation capabilities. The program will provide enabling technology for future manned, reconnaissance, and Uninhabited Combat Air Vehicle (UCAV) systems. The following describes the spectral calibration and flat fielding studies performed on a recently collected data set of the Dark HORSE 1 (DH1) visible hyperspectral sensor. The procedures used for spectral calibration and mapping of the sensor's focal plane array (FPA) are described and a real-time algorithm for corrective re-mapping of the image data is presented in detail. Also, procedures used for pre and post-flight flat fielding of the sensor's radiometric response are described with an emphasis on the potential advantages and disadvantages of each. Results of these efforts are discussed and appropriate conclusions are drawn.