DocumentCode :
859974
Title :
Spatio-temporal deconvolution of NDVI image sequences using independent component analysis
Author :
Lotsch, Alexander ; Friedl, Mark A. ; Pinzón, Jorge
Author_Institution :
Dept. of Geogr., Boston Univ., MA, USA
Volume :
41
Issue :
12
fYear :
2003
Firstpage :
2938
Lastpage :
2942
Abstract :
Independent component analysis (ICA) provides a powerful new method to spatially and temporally deconvolve image sequences into components that capture variability arising from independent physical sources. To do this, ICA uses information contained in higher order cross-moments of multivariate data. We use remotely sensed time series of the normalized difference vegetation index to illustrate the utility of this technique.
Keywords :
deconvolution; geophysical signal processing; image sequences; independent component analysis; vegetation mapping; NDVI image sequences; cross-moments; independent component analysis; multivariate data; normalized difference vegetation index; remotely sensed time series; spatio-temporal deconvolution; Data mining; Deconvolution; Image sequence analysis; Image sequences; Independent component analysis; Matrix decomposition; NASA; Principal component analysis; Spatial resolution; Vegetation mapping;
fLanguage :
English
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
Publisher :
ieee
ISSN :
0196-2892
Type :
jour
DOI :
10.1109/TGRS.2003.819868
Filename :
1260632
Link To Document :
بازگشت