Fringe Frequency Estimation Of Interferogram And Its Application In Interferometric Synthetic Aperture Radar

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. SAR image pair co-registration, phase noise filtering, and phase unwrapping are the key steps of InSAR processing. But the interferogram generated by a pair of SAR images may be contaminated by strong noise, and therefore unsuitable for further analysis in the spatial domain.In order to decrease the effects of noise, we propose to exploit the local frequency of interferometric fringes to improve the critical procedures in SAR interferometric processing. The main research work includes:1. Propose the coherence-based adaptive window for local fringe frequency estimation. This new method uses coherence as influencing factor for adaptive estimation, so that areas with low coherence are estimated using large window while those with high coherence using small one.2. Propose a residual linear phase trend removal algorithm based on instantaneous frequency estimation of interferogram. First, the instantaneous frequency of interferogram is estimated up to an accuracy of 0.1 fringes through the Chirp-Z transform. Then, the linear trend is simulated from the instantaneous frequency and been removed from the interferogram. Simulated and real data are exploited to validate the efficiency of this method. The results show that the method can more completely remove the linear trend in interferogram.3. Establish the interferometric fringes reconstruction and filtering method based on local frequency estimation. The proposed filter overcomes the shortcomings of breaking interferometric fringes when filtering by traditional filter. Simulated and real data over Etna areas are used to demonstrate effectiveness and advantages of this new filter. Experimental results indicate that the new filter can preserve the structure of fringes better while reduce more phase noise compared with the classical filter of Goldstein,.4. Investigate the phase unwrapping method based on local frequency estimation. This method can acquire the unwrapped phase directly without the filtering process. Experiments with both simulated and real data prove its feasibility...