Development
Plan
First year: Testbed will be established first. This will provide the resources for conducting metadata extraction research. The research activities include: construct comprehensive geographic databases containing a large volume of geospatial data, conduct experimentation of the feature extraction to compute multidimensional data representations of the images, and cluster the computed data.
Second year: A version of metaserver will be established. This will provide a platform to conduct re source selections. The research activities include: configure the templates in the metadatabase based on the multidimensional data, design selection approaches for the metasearch agent, and evaluate the effectiveness of the templates and other metadata in representing the clusters of image databases
Third
year: The establishment of a complete metasearch environment. The
research activities include: deploy the metaserver at the three sites,
conduct comprehensive examination of the performance of queries on
various features, including texture, color and text, using the proposed
selection approaches and the performance of the merging algorithm, develop a
set of software tools for building a metaserver on top of multidimensional
databases.
Expected
Accomplishments
The expected
accomplishments from this research project include data resources and software
tools:
Data
resources: A huge volume of geographic data will be collected and
classified based on the semantics. This asset will be accessible by users for
research and educational purposes.
Software tools: A set of software tools will be available for the use of establishing a metaserver environment which manages largescale multidimensional data resources.
