Hierarchical neural network approach to ocean colour extraction from remotely sensed imagery

Radiative transfer algorithms in combination with empirical formulae have been the most popular approach to the analysis of oceanic water types from remotely sensed satellite images of the Earth. These methods produce occasional errors created by unstable atmospheric components and disable monitoring of coastal zones. As the assumptions on sensor. Earth surface and atmospheric interaction with electromagnetic radiation are restraining, multi-spectral and fusion techniques based on the application of unsupervised neural networks can contribute to the improvement in ocean colour studies and enable analysis of complex wafer types. This paper presents the application of a hierarchy of self-organizing feature maps to feature extraction and differentiation of oceanic waters. The practical studies are performed on imagery captured around the Pacific Ocean by the ocean colour and temperature scanner on board of the Japanese Advanced Earth Observing Satellite.