A fast Lee filter algorithm based on Fourier transform

A fast Lee filter algorithm based on Fourier transform is proposed. The algorithm uses an optimized noise variance estimation method to achieve accurate estimation of SAR image noise variance. The convolution model realizes the calculation of the mean value and variance of the noise in the filter window, so as to realize the fast calculation of the Lee filter. The filtering algorithm has low complexity and high computational efficiency, and can realize real-time filtering processing of SAR images. Simulation experiments have verified the effectiveness and real-time performance of this method.