Translator Disclaimer
2 May 2012 Sparse and accurate high resolution SAR imaging
Author Affiliations +
We investigate the usage of an adaptive method, the Iterative Adaptive Approach (IAA), in combination with a maximum a posteriori (MAP) estimate to reconstruct high resolution SAR images that are both sparse and accurate. IAA is a nonparametric weighted least squares algorithm that is robust and user parameter-free. IAA has been shown to reconstruct SAR images with excellent side lobes suppression and high resolution enhancement. We first reconstruct the SAR images using IAA, and then we enforce sparsity by using MAP with a sparsity inducing prior. By coupling these two methods, we can produce a sparse and accurate high resolution image that are conducive for feature extractions and target classification applications. In addition, we show how IAA can be made computationally efficient without sacrificing accuracies, a desirable property for SAR applications where the size of the problems is quite large. We demonstrate the success of our approach using the Air Force Research Lab's "Gotcha Volumetric SAR Data Set Version 1.0" challenge dataset. Via the widely used FFT, individual vehicles contained in the scene are barely recognizable due to the poor resolution and high side lobe nature of FFT. However with our approach clear edges, boundaries, and textures of the vehicles are obtained.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Duc Vu, Kexin Zhao, William Rowe, and Jian Li "Sparse and accurate high resolution SAR imaging", Proc. SPIE 8394, Algorithms for Synthetic Aperture Radar Imagery XIX, 839407 (2 May 2012);

Back to Top