Neural Network Algorithms for Ocean Colour (NAOC)
The aim of NAOC is to provide advanced algorithms for better exploiting the information contained in new multi spectral ocean colour satellite observations and to obtain a better estimation of bio-optical components of the ocean. Attention will be focused on atmospheric corrections and retrieval of marine constituents. We will use neural methodology, which are well suited to model highly non-linear phenomena and for inverting noisy empirical data sets. Validation will be done on already existing or concomitant in-site observations. NAOC innovation is to bring together scientists having a background in marine and atmospheric sciences and applied mathematicians. Scientists, managers, policy makers will benefit of NAOC. Thanks to the improvement of ocean colour data, they will be able to better assess the role of ocean productivity in the management of the planet Earth and to answer to the interrogation of people for the evolution of the ocean.
An dem Projekt beteiligt sind: