Based On The Signal To Noise Ratio Of Insar Interferogram Phase Filtering Method

InSAR (Interferometric Synthetic Aperture Radar, InSAR) is an important technique for earth observation which has being developed rapidly for recent years and applied widely to topographic surveying, earthquake monitoring, volcano monitoring, etc. In InSAR processing, one of the most important step is phase unwrapping, which is directly restricted by the quality of the interferogram. Therefore, the interferometric phase images must be filtered before phase unwrapping to reduce the phase noise and to improve the accuracy of DEM and surface displacements detecting .Recent years, various filtering methods for SAR interferograms are presented in International Radar Interferometry field, including algorithms both in space and frequency domain. All of these filtering methods can reduce the phase noise of SAR interferograms in a certain extent. However, they have many disadvantages, e.g. the parameters of filtering are adopted by experience, which have great randomicity; the effect of phase noise restraining is not improved very much. Besides, inadequate adaption is the common limitation of them.Therefore, a new phase noises reduction method of InSAR interferogram based on Signal-to-Noise (SNR) is proposed in the dissertation in order to avoid above disadvantages. The new method uses SNR as influencing factor of self-adaptive filter and processes the phase with smooth effect. This new filtering method is a modification to the Goldstein interferogram filter. It makes the Goldstein's filter parameter a dependent on SNR, so that areas with low SNR are filtered more than that with high SNR. This improvement can decrease the losing of signal as well as reduce phase noise.In this paper, simulated and real interferometric data of Hong Kong region are used to demonstrate the effectiveness and advantages of the new method. The results of experiments indicate that compared with the classical and improved Goldstein filter, the new proposed filter of InSAR interferogram based on Signal-to-Noise (SNR) not only can reduce more phase noise in the interferogram, but also can preserve the smoothness and continuity of the stripe. The new method using SNR as the influencing factor of the filter for InSAR interferogram is first proposed in this dissertation. As SNR is the better-comprehensive evaluating criterion of image quality, the proposed filtering method can be considered to applied to other remote sensing image besides SAR interterogram.