希尔伯特—黄变换在遥感图像中的实例研究与应用展望

Study and Application of Hilbert-Huang Transform in Remote Sensing Images

希尔伯特-黄变换是一种处理非线性、非平稳数据的方法,目前该方法已在遥感图像处理中得到高度重视。文中在介绍希尔伯特-黄变换的基础上,首先通过对时间序列数据的不同频率信息进行分离,揭示了希尔伯特-黄变换方法对局部时间尺度的物理含义。通过希尔伯特-黄变换方法对不同空间尺度信息的分离,对SAR遥感图像滤波和SAR遥感图像信息提取开展了实例应用研究,展示了希尔伯特-黄变换方法可以通过遥感图像提供各种尺度信息现象的有效数据。最后讨论与分析了希尔伯特-黄变换方法的局限性与优势,并对其在遥感图像处理中的应用进行了总结和展望。

Hilbert-Huang transform is a method to process non-linear and non-stationary data, and has been highly valued in the field of remote sensing image processing. In this paper, after introducing the method of Hilbert-Huang transform, different frequency information is isolated through the examples of time series data, which reveals the physical meaning of local time scale. The separation of different spatial scale information is studied and discussed through the SAR image filtering and SAR image information extraction, showing the potential of the method of Hilbert-Huang transform to provide effective data in the study of remote sensing images. Finally, the advantages and limitations of the Hilbert-Huang transform are analyzed, and the prospect of the Hilbert-Huang transform is given in the application of remote sensing image processing.