Nearshore bathymetry estimation from SAR imagery based on the adaptive window guided by local wave direction

Nearshore bathymetry estimation is crucial for human marine activities. Synthetic Aperture Radar (SAR) plays an essential role in nearshore bathymetric mapping. In recent years, nearshore bathymetry estimation based on the variations in ocean wave characteristics from SAR imagery has garnered significant attention. However, existing Fast Fourier Transform-based methods are not accurate in estimating the wavelength when spectral peaks are not sharp, resulting in high bathymetric errors. To address this problem, we propose a new estimation method utilizing an adaptive windowing strategy. Specifically, the method adaptively sets the window angle based on the local wave direction of each sub-image and consequently generates multi-angle windows. The final wavelength is estimated using the Kalman filter, which leverages the differences of multi-angle windows to enhance bathymetric accuracy. Simulation and real SAR images are used to verify the effectiveness of this method. Compared with the traditional method, the proposed method can distinguish more subtle bathymetric changes and significantly reduce the bathymetric errors.