SAR Image Processing and Target Detection Based on 2D Singularity Power Spectrum in Time-frequency Domain

A method for SAR image processing and target detection based on two-dimensional singularity power spectrum (2D-SPS) in time-frequency domain is proposed in this paper. Firstly, the two-dimensional pseudo Wigner-Ville distribution (2D-PWVD) is carried out on the SAR image. Then, the two-dimensional singularity power spectrum estimation is performed on the local delay components of 2D-PWVD to obtain the three-dimensional singularity power spectral SAR (SPS-SAR) functions. Furthermore, a novel SAR target detection method based on SPS-SAR is introduced. The proposed methodology is tested on SAR images, both with and without weak targets, from the OpenSAR dataset. Experiment results show that the proposed method performs better than traditional method based on fractal theory, and can achieve higher detection probability in the complex scene with low SNR. It can provide more details for SAR image analysis and offer a new idea for SAR target detection.