Research Of InSAR Phase Unwrapping Algorithm At High Volume Subsidence Mining Area

Phase unwrapping is a key step in the InSAR technique.It is also the key and difficult point for InSAR technology research. Reasons such as image noise and undersampling that bring about low accuracy for phase unwrapping when mining subsidence monitoring, even affect the reliability of the InSAR mining subsidence monitoring results. This paper mainly studies the problem of phase unwrapping algorithm for InSAR technique, and the main work summarized as follows:1) Paper introduced the the basic principle of Goldstein branch cutting method, quality guidance method, least square method, and network flow algorithm.This paper utilize the actual SAR data and simulated data for experiment, and analysis the advantages and disadvantages of all kinds of phase unwrapping methods in mining subsidence monitoring, pointing out that various methods applicable to occasions.2) For the phenomenon of "isolated island" in Goldstein brantch cut algorithm ,this paper proposed an improved optimization branch cutting method. Pseudo coherence coefficient was introduced to the improved algorithm. Through the experiment and analysis, the improved method can effectively reduce the phenomenon of "isolated island", and improved the precision and reliablity of branch cutting method.3) For the problem of poorer phase unwrap that in high volume subsidence area, this paper proposed a new phase unwrapping algorithm which based on GVF-Snake model (gradient vector flow model). The New method firstly use GVF-Snake model to fringe boundary detection at discontinuous area, and then calculate the phase jumping count.At last,the algothrim utilize a rule about phase jumping count to generate new unwrapped phase. Paper also used three kinds of cases for experiment and analysis. Proved that the proposed method can solve the problem of low accuracy under the condition of high noise .4) In view of requirements of InSAR mining subsidence monitoring,paper used c++ and c# language to write the InSAR subsidence monitoring and analysis system, and integrated several phase unwrapping algorithms.