Coherent Integration with Multi-pass Synthetic Aperture Radar Images

Synthetic aperture radar (SAR) can be used to acquire high-resolution images of huge areas and has been widely used in planetary exploration. However, in earth-based SAR imaging of the moon, the effect of additive noise is not ignorable and the restricted temporal window for single observation limits the potential for image denoising through integration methods. To address this issue, a method to achieve coherent integration with multi-pass SAR images for additive noise removal is proposed. The proposed method achieves envelope alignment through sub-pixel level image registration and phase alignment by compensating the flat earth phase and topographic phase, thus satisfying the requirement of coherent integration. The signal-to-noise (SNR) gain resulting from this method is given by a function of the correlation. The effectiveness of the method and the correctness of the function are verified by the simulation experiments.