LAUSR

The backscatter of $X$ -band marine radar is contaminated by rain, which limits the estimation of wave parameters from radar images. A new method is proposed to correct the influence of rain on the gray level radar image, based on 1-D complex continuous wavelet transform. First, a suitable scale parameter is derived from the modulus of Gaussian wavelet coefficient, and then the wave crests are detected from the phase of the wavelet coefficient. Second, the gray levels of wave troughs are set to be a minimal constant, and the other gray levels are linearly interpolated for each radial. The method is validated by comparing the significant wave heights measured by buoy with those retrieved from $X$ -band marine radar image sequences, the correlation coefficient and the root-mean-square error between them are 0.85 and 0.42 m with the correction of rain, which increases 0.34 and reduces 0.34 m compared with those estimated without the correction, respectively. The recognition of rain from gray level radar image and the limitations of the method are discussed.