Dr. Michael Gertz
Department of Computer Science University of California, Davis

Title: GeoStreams: Managing Remotely-sensed Streaming Geospatial Image Data

Abstract: In the past few years, there has been an increasing interest in the adaptive and efficient processing of queries over continuous streams of data originating from network monitors, various types of sensors, and mobile devices. Most of the work, however, adopts the relational model and relational query operators as basis for processing stream data and queries over streams. Complex types of stream objects, such as spatio-temporal data or raster image data originating from the vast amount of remote-sensing equipment orbiting the Earth, have not been studied.

In this presentation, we give an overview of the GeoStreams project on adaptive query processing models and architectures for streaming geospatial image data. We present fundamental models and techniques for the real-time processing of remotely-sensed, streaming geospatial image data, with a particular focus on National Oceanic and Atmospheric Administration's (NOAA) Geostationary Operational Environmental Satellite (GOES). The proposed GeoStreams data and query model extends current approaches for processing continuous, multi-user queries on data streams to support the efficient processing of spatio-temporal image data. We discuss important operators on streaming geospatial image data, including spatial selections, band algebra operators, spatio-temporal aggregation, and spatial transformations (map projections), and outline how such operators are integrated into an adaptive query processing architecture. The primary application driving our research efforts is the calculation of spatially distributed daily reference evapotranspiration maps for the State of California at high spatial resolution (1 km2-16 km2) and in real-time.

This project is is a collaborative research effort between the Department of Computer Science and CalSpace Center of Excellence at the University of California, Davis, and is supported by an NSF ITR award.