Measurement of sea-surface velocities from sequential satellite images using the Hopfield neural network

The knowledge of ocean surface circulation is of major importance in various scientific applications including the understanding of global meteorological phenomena, resources exploitation, and containment of chemical spills. The cost and logistics involved in the in situ measurements of ocean surface velocity has led to the development of 2 classes of methods based either on surface feature tracking or on thermal equation inversion based on sequential satellite images. In this work, sea-surface feature tracking based on the Hopfield neural network was developed. The method is based on the minimisation of an energy function that represents the feature tracking problem. A Hopfield neural network is used to merge cross-correlation information with prior knowledge of sea-surface flows and image contextual information. This fusion of information, through the use of a simple energy function, is compared with the MCC method that does not use the convergence process. The use of contextual information and prior knowledge of flow enables the method to generate smooth and high resolution vector field showing all the details of the surface velocity distribution. The method can be used on various kind of images for tracking, and also find other applications in image registration and pattern recognition. The method was also tested on real satellite images. A set of 9 AVHRR thermal images of the coastal zone of California, along with a data set of coincident surface drifters positions were used to test the method. Results of the new analysis are compared with in-situ data and previous results using the same data and different methods.