HR: 0830h
AN: U41B-0013    [PDF]
TI: Applications of Bayesian Classification to Content-based Data Services
AU: Lynnes, C
EM: Christopher.S.Lynnes@nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Berrick, S
EM: Stephen.W.Berrick@nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Gerasimov, I
EM: gerasimo@daac.gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: * Gopalan, A
EM: gopalan@daac.gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Hua, X
EM: Xin-Min.Hua.1@gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: MacHarrie, P
EM: Peter.G.Macharrie.1@gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Ouzounov, D
EM: ouzounov@daac.gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Shen, S
EM: sshen@daac.gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Smith, P
EM: Peter.M.Smith@nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Yang, K
EM: kyang@daac.gsfc.nasa.gov
AF: NASA/GSFC, Code 902, Greenbelt, MD 20771 United States
AU: Wheeler, K
EM: Kevin.R.Wheeler@nasa.gov
AF: NASA/ARC, MS 269-2, Moffett Field, CA 94035 United States
AU: Curry, C
EM: ctc@email.arc.nasa.gov
AF: NASA/ARC, MS 269-2, Moffett Field, CA 94035 United States
AB: The high volume of Earth Observing System data has proven to be challenging to manage for data centers and users alike. At the Goddard Earth Sciences Distributed Active Archive Center (GES DAAC), about 1 TB of new data are archived each day. Distribution to users is also about 1 TB/day. A substantial portion of this distribution is MODIS calibrated radiance data, which has a wide variety of uses. However, much of the data is not useful for a particular user's needs: for example, ocean color users typically need oceanic pixels that are free of cloud and sun-glint. The GES DAAC is developing several experimental data services based on scene content for near-real-time MODIS calibrated radiance products (from Direct Readout stations). Content-based subsetting would allow distribution of, say, only clear pixels to the user if desired. Content-based subscriptions would distribute data to users only when they fit the user's usability criteria in their area of interest within the scene. Content-based cache management would retain more useful data on disk for easy online access. All of these services rely on a content characterization that must be performed immediately after production to be helpful. We use a simple Bayesian classification scheme, trained on other MODIS science products (such as cloudmask), to rapidly classify each pixel in the scene. The classification mask is then added to the calibrated radiance product to support the above content-based services.
DE: 1294 Instruments and techniques
SC: U
MN: 2003 Fall Meeting