Parallel Processing of High-Dimensional Remote Sensing Images Using Cluster Computer Architectures

Hyperspectral sensors represent the most advanced instruments currently available for remote sensing of the Earth. The high spatial and spectral resolution of the images supplied by systems like the Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS), developed by NASA Jet Propulsion Laboratory, allows their exploitation in diverse applications, such as detection and control of wildland fires and hazardous agents in water and atmosphere, detection of military targets and management of natural resources. Even though the above applications generally require a response in near real time, few solutions are currently available to provide fast and efficient processing of such high-dimensional image data sets. This is mainly due to the extremely high volume of data collected by hyperspectral sensors, which often limits their exploitation in analysis scenarios where the spatial and temporal requirements are very high. In this paper, we describe new parallel processing methodologies for hyperspectral image processing, based on neural architectures and morphological concepts. The computational performance of the proposed methods is demonstrated using real analysis scenarios based on the exploitation of AVIRIS data using two parallel computer systems, an SGI Origin 2000 multicomputer located at the Barcelona Supercomputing Center (BSC), and the Thunderhead Beowulf cluster at NASA’s Goddard Space Flight Center (NASA/GSFC).