Edzer Pebesma, Lubia Vinhas

to be presented at: 2009-10 Program on Space-time Analysis for Environmental Mapping, Epidemiology and Climate Change; Opening Tutorials & Workshop September 13-16, 2009

Change detection from imagery data is often performed conditional to time, meaning that time snapshots are classified independently and compared afterwards. This makes it hard to assess the statistical properties of the changes found. We will explore possibilities for assessing change from image time series based on unconditional, integral classifications of the series, and compare them to conditional approach. We will also look into the difference in requirements to ground truth data that arise from the two approaches. The application area is deforestation in the Amazon area.