节点文献

单轴分枝乔木分枝处的树皮纹理合成

Bark Texture Synthesis on Branches of Trees with Monopodial Branching

【作者】 崔颖

【导师】 曹卫群;

【作者基本信息】 北京林业大学 , 计算机应用技术, 2012, 硕士

【摘要】 树木的真实感建模与绘制不仅在于对树木整体形态的模拟,还在于对树木表面细节的模拟。树皮细节纹理的高真实感模拟对提高树木的真实感具有重要作用。由于大多数树木建模,特别是通过计算获得的树木几何数据所建立的模型,其主干与侧枝通常分别采用相互独立的曲面模型,难以通过传统的纹理合成方法实现分枝处树皮纹理的自然过渡,在很大程度上影响了树木模拟的真实感。而在自然界中不同种类的树木由于品种的差异,枝杈的生长和树皮纹理的过渡遵循不同规律。论文在基本的图割纹理合成方法基础上,提出了拼合曲面上的纹理合成方法,模拟单轴分枝乔木的主干与侧枝、侧枝与侧枝等相交位置的过渡树皮纹理。为了模拟自然生长的单轴分枝乔木分枝处树皮纹理各向异性的效果,论文将拼接曲面的合成区域划分成子区域,各子区域采用不同的合成方法。实验表明,该方法可以有效模拟单轴分枝乔木分枝处的树皮纹理,满足真实感绘制的需求。

【Abstract】 The realistic modeling and rendering of trees lies not only in the form of trees, but also in the simulation of the tree surface detail. The high realistic simulation of the bark detail texture plays an important role on improving the realism of trees. Since most tree modeling, in particular, the model obtained by computing geometry data of trees, its trunk and branches are usually modeled by independent surfaces, it is difficult to achieve the natural transition of bark texture at branches through traditional texture synthesis method, it affects the realistic simulation of trees on a large extent. Moreover, different types of trees in nature due to differences in varieties, growth of twigs and transition of bark texture follow different laws. This paper proposes a flattening surface texture synthesis method based on the basic graph cut texture synthesis method, simulates the transition bark texture on the intersected locations of trees with monopodial branching, such as trunk and branches, twigs and branches. In order to simulate the anisotropic effects of the bark texture on the branchings of natural growth trees with monopodial branching, this paper divides the synthetic surface region of flattened surface into the sub-regions, each sub-region using different synthesis methods. The experimental results show that our method can effectively simulate bark texture of branches of trees with monopodial branching, to meet the needs of realistic rendering.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络