各向异性多尺度自相似随机场与地形构建

Anisotropy Multi-scale Self-similarity Random Field and Terrain Construction

在长期自然环境的作用下,实际的地形表面是一个各向异性、非线性的2维场,且地形的自相似特性在整个尺度空间上并非恒定的,而是与度量的尺度范围相关。由于传统的自相似随机场的构建方法并不能准确反映地形的实际特性,因此提出一种首先应用多尺度自相似随机场构建各向同性地形表面,然后引入表征各向异性的结构滤波器,再通过谱变换构建具有各向异性的多尺度自相似地形的方法。实验结果表明,该地形构建方法不仅能够反映实际地形的自相似性依尺度而变化的特点,而且能构建山脉的走向。

Due to long period natural evolvement, real terrain surface is an anisotropic and nonlinear field. Moreover, terrain surface is not always so perfect that it keeps invariable self-similar characteristic in whole scale space. Traditional construction method based on isotropic single self-similar field has weaker ability to describe such complex surface. Consequently, multi-scale self-similar random field are used to construct isotropic multi-scale self-similar surface. And then, a structure filter, which transforms an isotropic field to an anisotropic field by spectrum filtering, is introduced to construct anisotropic multi-scale self-similar terrain. Simulation results show that presented approach not only constructs the orientation of mountain but also accord with varying degrees of roughness at different scales.