干涉合成孔径声纳局部相干法复图像配准

Complex Image Registration for Interferometric Synthetic Aperture Sonar with Local Coherence Method

  • 摘要: 复图像配准是干涉合成孔径声纳(interferometric synthetic aperture sonar, InSAS)信号处理中一个非常重要的处理步骤,直接影响到提取的干涉相位质量,关系到最终所重建的数字高程模型精度。为提升获取的缠绕相位质量,在充分分析InSAS成像特点的基础上,提出了一种基于局部相干法的InSAS缠绕相位获取方法。首先分析局部数据块内相位点距离差的变化特点,在此基础上论证了采用中心相位点距离差代替局部相邻相位点距离差的可行性,进一步导出了采用局部相干法提取缠绕相位的方法。仿真数据和实测数据实验结果表明:所提方法提取的缠绕相位质量显著优于经典复图像配准方法,残差点数量急剧减少,并且可有效克服传统复图像配准方法中全局偏移量拟合所带来的误差传播。

     

    Abstract:
    Objectives Complex image registration is a very important step during interferometric synthetic aperture sonar (InSAS) signal processing, and its performance is directly related to the accuracy of the final reconstructed digital elevation model. In order to improve the quality of the wrapped phase, a wrapped phase extraction method for InSAS based on local coherence method is proposed.
    Methods First, the variation characteristics of range difference in local phase data block are analyzed. According to the analysis, the feasibility of using the central phase point range difference to replace the local adjacent phase point range difference is proposed. Further, the method of extracting wrapped phase by coherent method is derived.
    Results The experimental results performed on simulated data and real data show that the quality of the extracted wrapped phase by the proposed method is significantly better than that of the classical method, the number of residues is sharply reduced, and the error propagation in the global fitting can be effectively overcome, which helps to verify the effectiveness of the proposed method.
    Conclusions The proposed method can effectively improve the quality of wapped phase extraction in the InSAS system.

     

/

返回文章
返回