光滑地表面毁伤检测方法研究

光滑地面毁伤程度评估非常重要，目前评估方法主要是基于光学图像，可是在恶劣的气候或环境条件下，很难获得光学图像，合成孔径雷达（SAR）克服了这个缺点. SAR图像是地表结构和电特征等地球物理参数的映射，通过SAR图像反演地表结构参数，可以推测出光滑地面毁伤程度. 本文以几何光学模型（GOM）为基础，建立神经网络反演模型，以获得光滑地表面受损后的粗糙度参数：表面均方根高度（σ）和表面相关长度（l），并进一步评估光滑地表面受损程度. 实验结果表明该方法可行.

A study of the damage detection method for smooth earth surface

Damage assessment of the smooth earth surface is very important, and at present the assessment methods are primarily based on optic images. It is rather difficult to obtain optic images under bad weather or environment, while Synthetic Aperture Radar (SAR) can overcome the shortcoming. SAR image is the mapping of some geophysical parameters, such as the earth surface structure and electricity features. Using SAR image to make inversion of the earths surface structure parameters can assess the damage of the smooth earth surface. A neural network inversion model based on the Geometry Optics Model (GOM) is proposed in this paper, and some roughness parameters about the damage of the smooth earth surface are obtained, including surface root mean square height (σ) and surface correlation length (\%l\%). Damage grade of the smooth earth surface can be assessed via these roughness parameters. The experimental results indicate that the method is feasible.