Comparative analysis for combination of unwrapping and de-noising of phase data with high speckle decorrelation noise

Unwrapping and de-noising are key processes for the restoration of phase data in the presence of high speckle decorrelation noise. Usually, there are two strategies to deal with noisy wrapped phase: de-noising before unwrapping, or unwrapping before de-noising. This paper aims at comparing the robustness and efficiency of the strategies. Six combinations which belong to different strategies are compared in this paper. Ten simulated phase maps with progressive noise standard deviations are generated based on the realistic speckle decorrelation noise to evaluate the performances of the approaches. The results of simulation show that de-noising with windowed Fourier transform filtering before unwrapping with the algorithm based on least-squares and iterations which belongs the first strategy has the best accuracy and acceptable computation speed for the restoration of high noisy phase data. Application of selected methods to experimental phase data from digital holography validated the analysis.