最小生成树相位解缠中冗余去除算法

Redundancy Removal Algorithm in Minimum Spanning Tree Phase Unwrapping for Interferometric SAR

目前,相位解缠的方法中以Goldste in提出的枝切法最为经典,它通过枝切线将残差点相互连接使得残差点电荷中和,在相位展开处理的积分中,积分路径不能穿过分支,从而限制误差的传播。但是由于枝切线连接策略的不合理,往往造成连成的枝切线过多、过长,甚至多条枝切线形成闭合区域,造成不能解缠的“死区”,特别是在残差点较多的情况,这种现象更为严重。本文在最小生成树原理的基础上,提出的枝切线冗余去除算法能简单、准确地去除枝切树中的冗余,并保证了相位解缠具有最好的效果。

Goldstein＇s branch-cut algorithm is now the most classical one of the known phase unwrapping algorithms, It connects residues with branch-cut lines to neutralize their charges so that the integral path can＇ t stride over the branches in the process of phase unwrapping, thus avoiding pervasion of error. However, due to improper connections, this always brings about excessive branch-cut lines, and even many lines form a closed ‘dead area＇ that can not be unwrapped, especially when many residues exist, This paper presents a removal redundancy algorithm in phase unwrapping based on minimum spanning tree theory, and its implementation is described in detail, It can accurately remove the redundancies of the branch-cut tree and ensure the effectiveness of phase unwrapping.