Landsat TM satellite imagery, topographic and climate ancillary data are used to build binary (forest/non-forest) and multiclass (forest canopy cover classes) classification models, trained using sample aerial photograph maps, across Victoria, Australia.
Therefore, the main objective of this study is to detect of thrips on the crop canopy images using SVM classification method.
Further analysis of land use and land cover classifications (pavement, building, canopy, grass, and water) on 3 scales (100 m, 625 m, and 1250 m) suggested the surrounding landscape had little impact on monk parakeet nest-site selection.
The canopy storage capacity and leaf wettability classification were determined at 12 water temperatures and a practical calculator to compute changes of S and contact angles of droplets was developed.
The parameters used in the RF classification were the GLAS canopy height, the SRTM roughness index, the difference between the GLAS highest elevations and the SRTM elevations and the difference between the GLAS centroid elevations and the SRTM elevations.
In a post-processing step the classification between open and closed canopy was revisited based on LiDAR pulse density so as to obtain a uniform classification of closed canopy for both sets of field plots.
The model was tested alongside simpler transmittance models that ignored shading and treated the canopy as either a binary (open or forested) classification or that used proximal canopy descriptors.
Accuracies of the moderate fire severity category for measures describing effects only to trees (percent change in canopy cover and basal area) indicated that the classifications were generally not much better than random.
This study is a novel exploration of ensemble diversity and its link to classification performance, applied to a multi-class canopy cover classification problem using random forests and multisource remote sensing and ancillary GIS data, across seven million hectares of diverse dry-sclerophyll dominated public forests in Victoria Australia.
