Attributed Scattering Center Extraction With Improved Orthogonal Matching Pursuit

The attributed scattering center extraction is aimed to solve a high-dimensional, and highly nonlinear problem, which generates high complexity and memory requirements. To alleviate this difficulty, this paper proposes an attributed scattering center extraction method with improved orthogonal matching pursuit (OMP). Local optimization orthogonal matching pursuit (LOOMP) and alternative optimization are utilized to estimate parameters and reduce the dictionary dimension. Experimental results show that the proposed method outperforms the existing frequency-domain method in terms of runtime and memory requirements.