Filippi, A. M.. University of South Carolina, filippi@ellie.cla.sc.edu
Carder, K. M.. University of South Florida, kcarder@monty.marine.usf.edu
Davis, C. O.. Naval Research Laboratory, davis@rsd.nrl.navy.mil

 
AQUATIC IMAGING SPECTROSCOPY ISSUES: NRL PHILLS SENSOR CALIBRATION USING A COASTAL TREE-SHADOW METHOD
 
A method for vicarious calibration of the Naval Research Laboratory (NRL) PHILLS (Portable Hyperspectral Imager for Low Light Spectroscopy), hyperspectral sensor was developed that takes advantage of shadows cast by trees and cliffs along the coast. PHILLS is a test bed for the NEMO (Naval EarthMap Observer). PHILLS was flown over the waters surrounding the San Juan Archipelago in August 1998 in concert with in situ data collection efforts from three boats. The PHILLS data were not fully calibrated when the data were acquired, precluding the use of standard atmospheric correction models. Thus, a sensor-calibration method appropriate for coastal areas was developed using pixel pairs in the shaded and adjacent waters, which utilized the sky radiance dominating the shaded pixel as a known calibration target. Since PHILLS is a push-broom sensor, each image column corresponding to a CCD element must be calibrated independently. To demonstrate the method, several transects extracted from imagery acquired over East Sound, Orcas Island, WA were calibrated and validated with field data, thus providing validation of an approach for acquiring calibration-adjustment data from the image itself. Calibration extension to the entire scene based upon a flat-fielding approach using a separate scene will also be reported.
 
Day: Tuesday, Feb. 2
Time: Poster
Location: Sweeney Center
 
Code: SS43TU1048S