2003a. Classification of ice types from Convair-580 data of Northumberland Strait, PEI

• Main Authors: Bernd Scheuchl, Ian Cumming, Irena Hajnsek
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: Canada; Sea ice
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.407.1042; http://sar.ece.ubc.ca/papers/ASAR_2003_Scheuchl.pdf

ABSTRACT / RESUME The discrimination between open water and sea ice, and/or between sub-classes of sea ice are challenges for SAR remote sensing. Compared to single channel SAR data, fully polarimetric data contain more physical information about the scattering processes, which can be used to increase the discrimination accuracy. Unsupervised polarimetric classification or segmentation approaches allow the discrimination between sea ice and open water, as well as between several classes of sea ice. However, the main problem lies in the physical interpretation of the obtained classes. Airborne polarimetric SAR data acquired with the Environment Canada CV-580 C-band SAR is classified using an automated Wishart classifier and class mean information is investigated. 1.