Download (PDF 1.03 MB): https://research.fs.usda.gov/download/treesearch/17321.pdf
Abstract
RADARSAT-1 synthetic aperture radar data, speckle reduction, and texture measures provided for separation among forest types within the Twin Cities metropolitan area, MN, USA. The highest transformed divergence values for 16-bit data resulted from speckle filtering while the highest values for 8-bit data resulted from the orthorectified image, before and after performing a histogram stretch. First-order texture derivatives of 8-bit data provided only modest separability, while secondorder texture derivatives of 8-bit data provided little, if any, separability. RADARSAT-1 imagery may provide for image separation among forest types provided that preexisting forest/nonforest land cover classifications are incorporated.
Parent Publication
Citation
Nelson, Mark D.; Ward, Kathleen T.; Bauer, Marvin E. 2009. Forest-cover-type separation using RADARSAT-1 synthetic aperture radar imagery. In: McRoberts, Ronald E.; Reams, Gregory A.; Van Deusen, Paul C.; McWilliams, William H., eds. Proceedings of the eighth annual forest inventory and analysis symposium; 2006 October 16-19; Monterey, CA. Gen. Tech. Report WO-79. Washington, DC: U.S. Department of Agriculture, Forest Service. 303-306.
