Statistics – Applications
Scientific paper
Jan 2002
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2002esasp.475..105s&link_type=abstract
In: Proceedings of the Third International Symposium on Retrieval of Bio- and Geophysical Parameters from SAR Data for Land Appl
Statistics
Applications
Agriculture, Land Cover
Scientific paper
The application of a knowledge-based classification approach has been studied for land-cover classification using polarimetric SAR acquired by the Danish L- and C-band polarimetric SAR (EMISAR). The advantage of knowledge-based and model-based techniques is that they are normally more robust than for instance supervised Bayes classification methods, because they are virtually independent of the specific SAR data used and the specific test site used. The classification scheme used is based on a scheme originally proposed by the University of Michigan, and it has been modified to cope with the different object classes in this case. The classification was evaluated using a large number of test areas, and for the broad classes forest, lake, spring crops, and winter crops very good results were obtained using acquisitions at both L- and C-band in April.
No associations
LandOfFree
Land-cover map information from polarimetric SAR using knowledge-based techniques does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Land-cover map information from polarimetric SAR using knowledge-based techniques, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Land-cover map information from polarimetric SAR using knowledge-based techniques will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1089421