The technologies developed by Amphitrite are based on the extensive oceanographic and meteorological expertise of its research team. Our origins tracks back to the French national research center CNRS and the Ecole Polytechnique laboratories, where we have developed, in the last few years, numerous tools to characterize the physical and dynamic properties of the ocean in real time. The AMEDA algorithm, developed by our team, is now know widely used by researchers, from various countries, to detect, characterize and track the trajectories of oceanic eddies either on standard altimetry products or numerical models.
The co-localisation of in-situ measurements, especially those of automatic argo profilers, with remote-sensing observations allows us to estimate part of the sub-surface structure of the ocean. Thanks to this methodology we were able to build a unique data-base of oceanic eddies: the DYNED-Atlas. This data-base characterizes the dynamical properties and the sub-surface structure of long-lived eddies for more than 18 years in two specific areas: the Mediterranean Sea and the Arabian Sea. This achievement was made possible due to a close collaboration of research laboratories with the company CLS and the Oceanography unit of the French Navy (Shom).
The real-time products now developed by Amphitrite are based on the updates of theses advanced processing tools .
During oceanographic campaigns, it is the high resolution visible images, in particular the Sea Surface Temperature, that allow to identify with a high accuracy the meanders of oceanic current or the exact position of eddies. These visible patterns of a complex dynamic flow are usually analyzed, on board, by a human expert. Inspired by the recent developments of AI computer vision, successfully used in medical imaging, we develop a deep learning procedure and a convolutional neural network architecture adapted to the analysis of Sea Surface Temperature images. The image below shows the dynamical eddy contours corresponding to the maximum speed of each eddy detected on the input SST image.