species-level differences in hyperspectral metrics among tropical rainforest trees as determined by a tree-based classifier了解高光谱指标的差异热带雨林树木的树分类器.pdf

Remote Sens. 2012, 4, 1820-1855; doi:10.3390/rs4061820 OPEN ACCESS Remote Sensing ISSN 2072-4292 /journal/remotesensing Article Species-Level Differences in Hyperspectral Metrics among Tropical Rainforest Trees as Determined by a Tree-Based Classifier Matthew L. Clark 1,* and Dar A. Roberts 2 1 Center for Interdisciplinary Geospatial Analysis, Department of Geography and Global Studies, Sonoma State University, Rohnert Park, CA 94928, USA 2 Department of Geography, University of California, Santa Barbara, Santa Barbara, CA 93106, USA; E-Mail: dar@ * Author to whom correspondence should be addressed; E-Mail: mateolclark@; Tel.: +1-707-664-2558; Fax: +1-707-664-3332. Received: 3 May 2012; in revised form: 14 June 2012 / Accepted: 14 June 2012 / Published: 18 June 2012 Abstract: This study explores a method to classify seven tropical rainforest tree species from full-range (400–2,500 nm) hyperspectral data acquired at tissue (leaf and bark), pixel and crown scales using laboratory and airborne sensors. Metrics that respond to vegetation chemistry and structure were derived using narrowband indices, derivative- and absorption- based techniques, and spectral mixture analysis. We then used the Random Forests tree- based classifier to discriminate species with minimally-correlated, importance-ranked metrics. At all scales, best overall accuracies were achieved with metrics derived from all four techniques and that targeted chemical and structural properties across the visible to shor