SeaBED is an integrated data set and testing infrastructure for researchers to validate subsurface aquatic remote sensing algorithms. The purpose behind developing SeaBED is to collect multiple levels of image, field, and laboratory data with which to validate physical models, inversion algorithms, feature extraction tools and classification methods for subsurface aquatic sensing using hyperspectral imaging. The focus of this testbed facility is a field site located on Enrique Reef in southwestern Puerto Rico. This field site, which includes a heterogeneous mixture of both coral reef and seagrass habitats, offers a well defined system for evaluating analysis techniques under natural environmental conditions. Data produced from the field site currently includes airborne, satellite, and field-level hyperspectral and multispectral images, in situ spectral signatures, and water bio-optical properties. This data provides a valuable combination of sensing imagery and fully characterized ground truth information for developing and validating remote sensing algorithms. A major accomplishment for SeaBED was the collection of high-resolution hyperspectral imagery and associated ground truth of the near shore reefs and coastal ecosystems in southwestern Puerto Rico in 2007. The mission included 1740 km² of imagery acquired at 4 m spatial resolution, with 110 km² enhanced coverage of four science areas at 1, 2, 4 and 8 m spatial resolutions to facilitate investigation of spatial scaling issues and the testing of subsurface unmixing algorithms. We present an overview of SeaBED and also describe particulars of the 2007 data collection campaign, including both image acquisition and field data collection efforts.

Date Published: 13 October 2008
PDF: 8 pages
Proc. SPIE 7105, Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2008, 71050C (13 October 2008); doi: 10.1117/12.803785

Published in SPIE Proceedings Vol. 7105:
Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2008
Charles R. Bostater Jr.; Stelios P. Mertikas; Xavier Neyt; Miguel Velez-Reyes, Editor(s)