Deriving morphographic classification for cellular network planning using SAR interferometry

This paper demonstrates that morphographic classification for cellular network planning can be produced using coherence and intensity information of SAR images. One SLC image pair of the Tandem Mission was used for the classification. The selected morphographic classes were water, open, forests, suburban and urban areas. As classification features, intensity, coherence and texture obtained from intensity and coherence images were applied. Since the water areas were difficult to separate from forest areas, the classification process was divided into two steps: in the first one, water areas were determined, and the second step concentrated in separating the other classes. The overall percentage of incorrectly classified cases were 16.2%.