New Geospatial Approaches to the Anthropological Sciences
edited by Robert L. Anemone and Glen C. Conroy
New Geospatial Approaches
                to the Anthropological Sciences

Chapter 7. Classifying Land Cover on Very High Resolution Drone-Acquired Orthomosaics


SA Wich, LP Koh, and Z Szantoi

Additional Chapter details here

Figures

Fig. 1. (A) The location of the study area on Sumatra; (B) an orthomosaic of the study site; (C) an enlarged part of the study area; (D) an enlarged part after segmentation step.

Fig. 2. Land-cover class results from pixel-based analyses for (A) the full set and (B) the combined set.

Fig. 3. Land-cover class results from object-based analyses for (A) the full set and (B) the combined set.

Fig. 4. Producer and user accuracies for pixel-based (Maximum Likelihood) and object-based classifications (Support Vector Machine) for the combined and full sets. Panels (a) and (b) show the full set of classes, and panels (c) and (d) show the combined set of classes.