Consolidation of a pixel-based classification using neighborhood information

TitleConsolidation of a pixel-based classification using neighborhood information
Publication TypeConference Paper
Year of Publication2003
AuthorsScheuchl, B., I. Cumming, and I. Hajnsek
Conference NameGeoscience and Remote Sensing Symposium, 2003. IGARSS '03. Proceedings. 2003 IEEE International
Pagination449 - 451 vol.1
Date Publishedjul.
KeywordsAIRSAR ice data, classification confidence measure, geophysical signal processing, image classification, image consolidation, image speckle, neighborhood information, pixel-based classification, radar polarimetry, remote sensing by radar, smoothing, synthetic aperture radar, terrain mapping

A two-step consolidation process is presented that reduces the effect of speckle in the classification result. The degree of consolidation or smoothing can be adjusted using a classification confidence measure and a threshold for the amount of neighborhood agreement needed. Results from AIRSAR ice data shown are compared to a classification result form pre-filtered data.


a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2021 The University of British Columbia