De-noising method of InSAR data based on empirical mode decomposition and land deformation monitoring application

Abstract

The applicable choice of filters for InSAR is one of the key procedures, which is associated with the quality of interferogram. The data of InSAR interferogram was decomposed by empirical mode decomposition (EMD). A given signal was decomposed into different Intrinsic Mode Functions (IMFs) filled with the condition. Further investigation of the algorithm is demonstrated below with regard to the multi-resolution standpoint. Empirical mode decomposition includes two operators. The IMF calculation operator and residual calculation operator define the process of similar to high frequency and low frequency filters. Then, the multi-solution structure is realized by decomposing the low frequency step by step. Therefore, the filtered noise-related IMFs together with the other IMFs can be used to restructure the denoised signal. The processing result has confirmed this method feasibility. Comparing the empirical mode decomposition with the general methods, such as median filter, Lee filter, Goldstein filter, using the quantitative evaluation index, i.e., standard deviation (STD) and equivalent number of looks (ENL), the result shows that empirical mode decomposition is powerful to interferogram speckle noise suppression and residues reduction, as well as it can be preserved details information. The method proposed can improve the accuracy of interferometric products.