Change Detection with SAR Images Based on Radon Transform and Jeffrey Divergence

Focusing on the change detection with multitemporal Synthetic Aperture Radar (SAR) images, this paper presents a new approach based on the comparison of the density of the projections produced by Radon transform. The projections include the structure information, which helps when the local statistical distribution does not change. Edgeworth approach is used to fit the statistical distribution model of the projections. Jeffrey divergence is proposed as a measurement of the difference between two densities for that it is numerically stable and robust with respect to noise. This approach is demonstrated feasible according to the processing test using real satellite SAR images.