<p class="Default" style="text-align:justify;"><span style="font-size:11pt;">This study examined the capability of high-resolution imageries for identifying tree species. The IKONOS and CASI (Compact Airborne Spectrographic Imager) data were examined to digitally identify 20 tree species and estimating stand density. The numerical taxonomy using nearest neighbor hierarchical classification method was applied to cluster the spectral reflectance of those species of interest. Although the panchromatic band of IKONOS and CASI have the same spatial resolution, the study shown that CASI provided better performance than IKONOS in discriminating 20 tree species of interest. The finer spectral and spatial resolution of CASI significantly improved the quantitative discrimination ability. Inversely, the IKONOS imagery was fail to digitally identify tree species. However, the study shows that both the IKONOS and CASI images are capable to be used to estimate the stand density. To get a better result of discriminating 20 species using CASI image, the number of bands hould be used more than eight bands. Otherwise, some "inseparable" class pairs could exist. </span></p> <p class="MsoNormal" style="text-align:justify;">Keywords: CASI, IKONOS, Separabilitas, Klaster</p>
by: I Nengah Surati Jaya
Published: (2011-01-01)