Research Of Improving The SAR Imaging Quality Based On Sparse Regularization

Resolution and side lobe level are important indexes to measure the quality of a radar image.The quality of a radar image determines its application range and application effect.The higher the resolution,the more conductive to distinguish smaller targets.The lower the side lobe,the lower the possibility of false targets appear in the radar image.Whether it is in the military field or in the civilian field,improve the quality of radar image has a very important significance.Traditional radar imaging method has many limitations,and the imaging quality of the radar image obtained by traditional radar imaging method is limited.In this paper,based on the analysis of radar imaging model and sparse prior,the sparse constraint regularization model is established,and two sparse reconstruction algorithms are proposed to improve the quality of radar images.The first is the sparse reconstruction method which is based on the Fourier dictionary.This method is conducted in the complex frequency domain.By using this method,we can get a radar image with higher resolution and lower side lobe.However,the drawback is that the computational time by using this method is very large.It is very difficult to meet the real-time requirements.The second method is based on the unit dictionary.The method should be conducted in the complex image domain.By using this method,we can get a radar image with the same resolution and the lower side lobe level.