Teaching while selecting images for satellite-based forest mapping

Satellite images are increasingly being used to monitor environmental temporal changes. The general approach is to compare old images to recent ones acquired from a satellite in order to detect changes that occurred during the period between which these images were taken. An important step in this overall process is the acquisition of image data that will best allow assessing the temporal changes. This acquisition requires some expertise in order to select images from satellite sensors that use the appropriate spectral band for a forest application. For instance, a sensor suitable for classifying urban images will not necessarily be appropriate for forest mapping, because buildings reflect electromagnetic waves differently from trees and hence show a different spectrum with the same sensor. In this paper, we present an emerging Image Data Selection Assistant (IDSA) that uses an expert system combined with intelligent tutoring to help users in choosing images for updating forest maps, while at the same time teaching them how to best select images depending on the task.