This is often the main purpose of digital image analysis--detecting changes in a landscape over time. This may be done for environmental purposes (e.g., examining deforestation in the Amazon) or for taxation purposes (e.g., looking for changes to properties, such as new pools in a backyard). Regardless of the reason for the analysis, there are a variety of approaches commonly taken. A very good overview of the different approaches is provided by the Center for Biodiversity and Conservation at the American Museum of Natural History. Today's lecture notes are taken directly from the Image interpretation and classification notes on land cover change methods written by Ned Horning.

A brief discussion of Idrisi's Land Change Modeler is provided here. A YouTube video demonstrating image extraction (mapping polygonal change over time). A complete set of YouTube videos that cover a wide range of remote sensing topics (from unsupervised classification to supervised classification to evaluating classification results), produced by Riccardo Klingers. An interesting tutorial on how to use GIMP to do some simple change detection. A scientific review of methods used in change detection (don't worry about following the math; Parts I and II a brief summary how change detection can be approached). A link to the thesis (Change Detection in Land Use and Land Cover using Remote Sensing Data and GIS) that I used in today's lecture (showing landuse change in Nigeria)

Text: Nothing is particularly relevant.

Overheads:

• One to a page
• Three to a page