Target and change detection in synthetic aperture radar sensing of urban structures

Cross-correlation based image subtraction for detection of targets within an urban structure from synthetic aperture radar images is presented. In surveillance operations requiring re-imaging of the same scene, small displacements in array element locations may cause large phase and amplitude offsets. Subsequently, clutter will not cancel out, but will rather persist when subtracting two images, one without the target and the other with the target present. We propose a correlation-based robust technique that mitigates image offsets and rotations and is applicable to synthetic aperture radar urban sensing where the statistics are highly non-Gaussian. Simulation results demonstrating the effectiveness of the proposed technique are also presented.