High Resolution Imagery Retrieval for Terrestrial Vegetation and Land Use Study

Study results of content-based retrieval from high resolution image data base (HRIDB) are presented. With the launch of each remote sensing satellite, the data is increasing exponentially both in content and usage. Multiple terabytes of HRIDB are being collected by majority of nations across the globe. It raises question how to retrieve, manage and make best use of the HRIDB information.
The handling of terrestrial vegetation imageries is specific due more or less structured sketch of the patterns. The HRIDB contains different types of terrestrial vegetation and land use data and contributes towards the extraction of sketches of different patterns. The study explores the grey level spatial dependence and relationship of high resolution patterns. A query result is based upon meaningful combinations of visual features and interaction structure of homogeneous patterns.