HR: 1340h
AN: A33C-1419 [Abstracts]
TI: Increasing MODIS aerosol retrievals with image inpainting for 3D-AQS
AU: * Wimmers, A J
EM: wimmers@ssec.wisc.edu
AF: Cooperative Institute for Meteorological Satellite Studies (CIMSS)
University of Wisconsin - Madison, 1225 W. Dayton St., Madison, WI 53706,
AB:
The MODIS retrieval of aerosol optical depth (AOD) is an essential component of the 3D-AQS (3-Dimensional Air
Quality System) collaborative project. In 3D-AQS, satellite, ground and lidar observations of aerosol are collected
and displayed in real-time for state and local air quality applications. Although AOD is effective at displaying major
aerosol events over the United States, large data gaps due to cloudiness, sun glint and high albedo can impede
the interpretability of the AOD imagery. This research explores the application of "inpainting" techniques to fill in
data gaps with interpolated values. Unlike conventional interpolation techniques, inpainting preserves edge
boundaries in all directions, which minimizes the appearance of interpolation artifacts. Results will show the
upper limits of the data gap size that can be effectively restored with inpainting, as well as the relative accuracy of
inpainted values.
DE: 0305 Aerosols and particles (0345, 4801, 4906)
DE: 0345 Pollution: urban and regional (0305, 0478, 4251)
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting