Automatic extraction of buildings and modeling of their spatial arrangements provide essential information for ur- ban planning, development, and monitoring applications.In this work we propose a method on modeling urbanization using spatial building patterns. Building detection is done using Bayesian classification of multi-spectral informa- tion. The individual buildings are used as textural primitives, and co-occurrence-based spatial domain features and Fourier spectrum-based frequency domain features are used to model their repetitiveness and periodicity at particular orientations. These features are used to quantify the building patterns and classify neighborhoods as organized (reg- ular) and unorganized (irregular). Experiments with high resolution Ikonos imagery show that the proposed technique can be used for automatic segmentation of urban scenes and extract valuable information about urban growth.

DATE: 26 March, 2007, Monday@ 16:20
PLACE: EA 409