Quantitative estimation of settlement density and limits based on textural measurements

This paper explores a new method for quantitative estimation of building density and settlement edges based on textural measurements. The evaluation employs an ordered multi-scale, linear regression scheme in which scale corresponds to spatial resolution and is represented by a kernel. The kernel size in the experiment described, ranges from 30 to 1000 meters for a test area of approximately 10.6×26.6 km over the city of Sana'a, Yemen. The input data is a very high resolution (VHR) satellite image, courtesy of QuickBird, and the reference data is the cadastral map of the city at scale 1:10.000. Computing the goodness of fit measure R
²
between the image-derived measurements and the reference data shows an overall performance for the settlement density of 80% at scale s = 200 m and 90% at s = 1 km.