A Study for Near-Shore Characterization Using High-Resolution Hyperspectral and Multispectral Images

Our goal is the development of spectral analysis tools that fully exploit the information content in multi/hyper-spectral imagery, particularly as it applies to remote sensing of ocean color and the extraction of bathymetry, water visibility and bottom type information. The aim of this study is the validation of standard bathymetry algorithms by comparing results from different sensors with a reference data set of in situ samples and the improvement of the resolution and estimation error by using the fusion of complementary sources. Multi-sensor fusion of satellite and in situ calibration measurements can be used to assist in performing Rapid Environmental Assessment (REA) of coastal area. The greater part of the data used here comes from a NATO Battlespace Preparation field experiment that was performed in April/May 2007 along the coast of Tuscany (BP07). During this campaign calibration taps and in situ sensors were employed to measure several bio-optical and physical parameters that allow integration with satellite-derived products to improve the final accuracy.