Monitoring land cover change detection with remote sensing methods in magnesia prefecture in Greece

Abstract

The use of methodological tools of Remote Sensing such as spatial comparisons and classifications of satellite images results to the generation of thematic maps showing land cover change. This work originates from the LIFE96ENV/GR/580 E.U. project for the Magnesia prefecture in Central Greece. Three different methods of supervised classification (minimum to distance, parallelepiped and maximum likelihood) are applied. A comparison between them shows that the maximum likelihood method provides the best accuracy assessment. at about 77%. Finally, the change detection maps for each of the 17 land-cover categories are produced. Conclusively. two kinds of connotation are shown: 1) The methodological. referring to the processes, the applied methodology and the spatial and qualitative results (error matrix tables. accuracy assessment) of the change detection techniques that were used, and 2) the thematic. resulting to the creation of 33 different thematic maps (ranging from the map of resampling LANDSAT and SPOT satellite data to the land cover change detection maps).