A computational mapping engine portal for accessing geolibraries

Computational portals can provide user customized front ends to very large geospatial image data archives. The computational portals differs from most portal technology in that it not only provides the user the interface for selecting an area and data type of interest, it can also permit the user the ability to manipulate and transform imagery data from a variety of sensors into a format most efficient for the desired end use. Portal technology allows tasks to be saved and configured as single process or as a sequence of chained processes to conduct advanced analysis on selected data for a defined area of interest. The approach offers advantages over conventional methods because it provides methods of generating and delivering data and information products the computational intricacies of which are totally transparent to the user. Mississippi State University has begun developing a computational portal for accessing a geospatial data library (geolibrary). The Computational Mapping Engine (CME) Portal at MSU began as a conventional interactive geospatial image archive and has successfully implemented computational portal technologies to address challenges in loading, storing, distributing, archiving, transforming, manipulating, and fusing remote sensing image data to facilitate the effective use of these data in research and in real-time decision making processes. This paper presents a general CME Portal architecture that can be used to develop and implement computational portals for accessing and processing remote sensing image from local or distributed geolibraries.