Title :
Evaluation of Convair-580 and Simulated Radarsat-2 Polarimetric SAR for Forest Change Detection
Author :
Goodenough, David G. ; Chen, Hao ; Dyk, Andrew
Author_Institution :
Pacific Forestry Centre, Natural Resources Canada, Victoria, BC
fDate :
July 31 2006-Aug. 4 2006
Abstract :
Radarsat-2 is a significant advancement in technological capability over Radarsat-1 in terms of better spatial resolution and polarimetric parameters. To determine the effectiveness of C-band polarimetry of Radarsat-2 data for applications of forest typing, structure recognition and disturbance detection, C-band polarimetric data from Environment Canada´s Convair-580 were acquired. We also obtained simulated Radarsat-2 data, a degraded version of the Convair-580 data. Polarimetric techniques, such as polarimetric filtering and decompositions, were performed on these data sets. The underlying backscatter phenomena were examined and the results were compared. Historical fire disturbance patterns were detected clearly with the Convair-580 data by using the Entropy- alpha decomposition technique. Moreover, the combination of the surface scattering and volume scattering decompositions showed results in obtaining clearcut information in forested areas. By using the simulated Radarsat-2 data, we gained an insight into what could be expected from real Radarsat-2 quad-pol products. Even though the simulated Radarsat-2 data had more noise than Convair-580 data, the fire scars in the same area were still detectable.
Keywords :
airborne radar; forestry; geophysical signal processing; geophysical techniques; radar polarimetry; radar signal processing; remote sensing by radar; spaceborne radar; synthetic aperture radar; C band polarimetry; Convair-580 polarimetric SAR; Environment Canada; Radarsat-2 polarimetric SAR; entropy-alpha decomposition technique; forest change detection; forest disturbance detection; forest structure recognition; forest typing; historical fire disturbance patterns; polarimetric decomposition; polarimetric filtering; radar backscatter phenomena; Computational modeling; Computer science; Computer simulation; Fires; Forestry; Polarimetry; Polarization; Radar detection; Radar scattering; Spatial resolution;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2006. IGARSS 2006. IEEE International Conference on
Conference_Location :
Denver, CO
Print_ISBN :
0-7803-9510-7
DOI :
10.1109/IGARSS.2006.462