Land cover dynamics classification using multi-temporal spectral indices from satellite image time series

Satellite images of medium resolution are often part of historical archives, which makes them ideally suited to study urban growth. Such archives enable creation of satellite image time series (SITS). SITS represent an amount of information greater far greater than individual images, but are complex and difficult to interpret. In this paper we develop Landsat TM SITS centered over Bucharest, we create temporal signatures based on spectral indices, and we show possible uses for change detection based on these temporal signatures.