InSAR Phase Unwrapping Method Based On Quality Map Segmentation

Quality guidance method is a typical algorithm of path following algorithm,which is widely used,compared with other path tracking methods,it uses an important auxiliary information: phase quality map.With the help of phase quality map,it is possible to obtain the mass values of each point in the phase diagram,and then phase unwrapping from the high quality region to the low quality region.Under the correct guidance of quality map,the quality guidance method can obtain higher accuracy,but it is very dependent on the accuracy of the quality map,so it is necessary to have a reliable phase quality map in order to obtain high precision phase unwrapping results.The traditional quality guidance method directly uses the numerical distribution of the quality map to select the different priority paths,extend from high quality area to cover all areas,in the process of solving the model,often affected by the instability of the numerical values in the mass map,easily lead to the wrong path,resulting in large areas of error settlement results,the accuracy of the whole interferogram is decreased.This article is mainly to improve the quality guidance law,for the lack of quality guidance law,a phase unwrapping method based on mass image segmentation is proposed.The main research contents and results are as follows:1: Combine quality map with residual point information,determination of residual mass values using phase quality map,select the appropriate segmentation threshold,using the appropriate segmentation method,the original coherence data is divided into high and low quality areas,will easily lead to error of the residual all included in the low quality region.2: If use the same method to solve the high and low quality areas,will inevitably lead to error,so according to the various phase characteristics around the algorithm,using different algorithms of the two areas of unwrapping respectively,the quality value of branch cut with high quality good areas by unwrapping;and in the quality value of low quality area is not good,the residual points by combining results of unwrapping wrapped using the proposed method solution of high quality region,so it can ensure that the error propagation solutions of low quality cannot be wrapped around the area,improving the unwrapping accuracy of the overall solution bound.One of the key technical problems in this method is how to choose the appropriate segmentation method to the original coherent data segmentation,so as to maximize the high and low quality segmentation,reduce the error;another problem is how to choose the appropriate phase unwrapping method for the low quality region to reduce the error to the minimum.In the experimental data,this paper uses a series of experimental data,including simulation data and real data,several typical phase unwrapping algorithms are compared with the algorithms proposed in this paper.In the aspect of error evaluation,this paper uses the mean square error,the calculation time and the back winding mean variance to evaluate the performance of the solution.The experimental results show that the advantages of this method are obvious,verify the effectiveness of the phase unwrapping algorithm based on quality graph segmentation in reducing error transfer.