Near Range Radar Image Reconstruction Algorithm by Weighted Envelopes Transformation

Abstract

We proposed a methodology to apply inverse boundary scattering transform to process complex radar data with minimum imaging artifacts. Unlike other boundary extraction-based methods, such as shape estimation algorithm based on boundary scattering transform and extraction of directly scattered waves, the developed approach does not rely on peaks of signals, does not require to connect points to quasi wavefronts, and it is numerically stable and robust. Instead of quasi wavefronts extraction, the method process huge amount of the wave-tracking lines, match the weight values to them, and apply multistage filtration based on the first and second derivatives. It makes the developed method to be able to be used for processing cluttered data like GPR. Using the experimentally obtained GPR profile, we demonstrated that the method allows separating the buried objects from clutter, noise, and imaging artifacts. Compared to conventional diffraction stacking method, weighted envelopes transformation results show good clearness of the targets, which makes the profiles easy to interpret. A number of filtration parameters over huge number of extracted features provide good flexibility and wide applicability of our approach.