基于修正嵌入式容积卡尔曼滤波的相位展开算法  

A Fresh Phase Unwrapping Method Based on Modified Embedded Cubature Kalman Filter

作　　者：代高兴 谢先明[1] 

机构地区：[1]桂林电子科技大学信息与通信学院,广西桂林541004

出　　处：《测绘学报》2017年第12期1998-2005,共8页Acta Geodaetica et Cartographica Sinica

基　　金：国家自然科学基金(41661092;61461011);广西自然科学基金重点项目(2016GXNSFDA380009);广西自然科学基金(2014GXNSFBA118273);广西无线宽带通信与信号处理重点实验室2014/2015年主任基金(GXKL061503);2017年桂林电子科技大学研究生科研创新项目(2017YJCX23)~~

摘　　要：针对干涉图展开问题,将Levenberg-Marquardt方法修正后的嵌入式容积卡尔曼滤波器应用于缠绕相位图像的展开中,结合基于修正矩阵束模型的局部相位梯度估计算法以及量化跟踪策略,提出一种基于修正嵌入式容积卡尔曼滤波的相位展开算法。该算法不仅可以精确和快速地展开缠绕像元,还可以在展开缠绕像元的同时抑制相位噪声,降低前置预滤波器的难度与复杂度,甚至可以在处理受噪声污染不严重的干涉图时免去前置预滤波处理步骤。试验结果表明本文算法具有较高的效率和良好的稳键性。A phase unwrapping algorithm is proposed that based on modified embedded cubature Kalman filter for interferometric fringes.The algorithm is the result through combining an embedded cubature Kalman filter modified by a Levenberg-Marquardt method,a robust phase gradient estimator based on amended matrix pencil model,and a quantization path-following strategy.This method can accurately estimate unambiguous unwrapped phase of interferometric fringes by applying the modified embedded cubature Kalman filter to perform phase unwrapping and noise suppress simultaneously along the path routed by the quantization path-following strategy,which is beneficial to simplify the complexity and the difficulty of pre-filtering procedure followed by phase unwrapping procedure,and even can remove the pre-filtering procedure.Results obtained with synthetic data and real data show more acceptable solutions with the proposed method,compared with some of the most used algorithms.

关 键 词：相位展开 嵌入式容积卡尔曼滤波器 Levenberg-Marquardt方法 量化跟踪策略 

分 类 号：P236[天文地球—摄影测量与遥感]