Research on the Construction of Data Management System of Massive Satellite Images

摘要

With the rapid development of aerospace remote sensing technologies, enormous challenges have been brought to the storage organization and high-effective management of the massive satellite images. By analyzing the big-data characteristics of the satellite images, this paper introduced and constructed the distributed database based on cluster storage system with regard to the selection of storage system types, the design of database models and the development of the management system. Focusing on solving the problems, such as the low-speed of imagery big data archiving, inadequate performance of query service and the inefficiency of statistics and analysis, this paper conducted the research on the multi-processor parallel archiving, relation queries on JSON and fast statistics analysis method of grid imagery coverage, and achieved the massive satellite image management with the daily data increment of terabytes and the annual increment of petabytes of data.