SAR interferometrie phase filtering based on wavelet transform and local frequency estimation

A novel approach combining the local frequency estimation with wavelet transform is presented to reduce interferometric phase noise for InSAR. First, the maximum likelihood estimator is used to obtain the frequency range of the noisy interferogram. Then, the wavelet transform is employed to obtain the wavelet coefficients of the real and imaginary parts of the complex interferogram. For the wavelet coefficients within the estimated frequency range and that out of the range, the NeighShrink and VisuShrink methods are employed respectively to shrink them. As a result, the noise can be effectively filtered without the loss of detailed information of the interferogram based on the advantages of the two shrinkage methods. The performance of noise reduction and fringe preservation is verified by the experiments with real interferogram.