系统仿真学报 ›› 2017, Vol. 29 ›› Issue (3): 516-523.doi: 10.16182/j.issn1004731x.joss.201703007

• 仿真系统与技术 • 上一篇    下一篇

基于GPU的坡面细沟发育过程可视化模拟

龙满生1,2   

  1. 1.井冈山大学电子与信息工程学院,江西 吉安 343009;
    2.流域生态与地理环境监测国家测绘地理信息局重点实验室, 江西 吉安 343009
  • 收稿日期:2015-06-01 修回日期:2015-07-21 出版日期:2017-03-08 发布日期:2020-06-02
  • 作者简介:龙满生(1977-),男,江西万载,博士,副教授,研究方向为图像分析、虚拟仿真。
  • 基金资助:
    国家自然科学基金(41561065),江西省自然科学基金(20151BAB203038)

Visual Simulation of Eroded Rill Evolution Process on Graphics Processing Unit

Long Mansheng1,2   

  1. 1. School of Electronics and Information Engineering, Jinggangshan University, Ji'an 343009, China;
    2. Key Laboratory of Watershed Ecology and Geographical Environment Monitoring, NASG, Ji'an 343009, China
  • Received:2015-06-01 Revised:2015-07-21 Online:2017-03-08 Published:2020-06-02

摘要: 为了快速表现降雨条件下坡面水蚀微地貌的动态发育过程,构建了基于元胞自动机的坡面水蚀模型,提出了基于GPU的可视化模拟算法。模型包括降雨、入渗、产流、侵蚀、沉积和输沙等过程,通过对元胞应用简单的水沙交换规则来实现产流产沙的同步演化。利用Perlin噪声来模拟微地貌和降雨等侵蚀因子的空间分异性与随机性,增强了坡面细沟发育的真实性。结果表明,该模型能够有效表现坡面细沟发育过程的分形特征,借助GPU可以大大提高坡面水蚀可视化模拟的实时性。

关键词: 坡面侵蚀, 细沟发育, 元胞自动机, 可视化模拟, 图形处理器

Abstract: A slope erosion model based on CA (Cellular Automata) and a corresponding simulation algorithm based on GPU (Graphics Processing Unit) were proposed to visualize rapidly the dynamic process of water erosion and microrelief development on hillslope under rainfall events. The model is composed of processes such as rainfall, infiltration, runoff, erosion, deposition, and sediment transport. At each time step, water and sediment were exchanged among adjacent cells observing sediment mass conservation law and flow continuity equation. All cells update their runoff and sediment simultaneously complying with the above simple rules so as to realize the evolution of hillslope microrelief. Perlin noise was superimposed on the slope surface to model the spatial variation and randomness of erosion factors such as microrelief and rainfall, which enhanced the visual effect of rill development. The experiment results show that the slope erosion model based on CA can effectively characterize the rill development process on hillslope, and GPU can be used to greatly improve the real-time performance of slope erosion visualization.

Key words: slope erosion, rill evolution, cellular automaton, visual simulation, graphics processing unit

中图分类号: