基于改进粒子群算法的二维相位解缠方法

摘要

Branch-cut is one of the most important approaches for the phase unwrapping, which is a key process for the data elevation in the interferometric syn thetic aperture radar (InSAR). Generally, the shorter the length of the branch-cut, the better the result of phase unwrapping. Therefore, how to determine the length of the branch cut becomes very important in this kind of branch-cut-based approaches. In this paper, a novel method for optimizing the length of the branch-cut is presented based on particle swarm optimization(PSO) algorithm, which has already been successfully applied to find the shortest path in the traveling salesman problem (TSP) theory. In order to remedy the local convergence problem, the mutation operator of the genetic algorithm is further introduced and combined with the PSO al gorithm. Compared to the branch-cut of Goldstein, this method can effectively reduce the length of branch-cut, and also avoid the "isolated island phenomenon" in the unwrapping process. Finally, the effectiveness and feasibility of the proposed method have been verified through both the simulated and experimental data.%相位解缠是干涉合成孔径雷达（InSAR）干涉数据处理的关键步骤之一，枝切法作为一种经典的相位解缠算法在相位解缠过程中，建立的枝切线长度越短相位的解缠效果越好，枝切线该如何建立至关重要。在旅行商问题（TSP）理论求解最短路径的思想下，一种利用粒子群优化算法建立最短枝切线的方法被提出，该方法在基本粒子群优化算法中引入变异算子对粒子群算法容易陷入局部收敛的缺陷进行了改进。相对于Goldstein枝切法而言，该方法能够更有效的缩短枝切线的长度，避免解缠过程中“孤岛现象”的发生。通过仿真和实测数据验证了该方法的可行性与有效性。