一种大曲率线状实体的三维可视化方法  

A 3D visualization method for high-curvature linear entity

作　　者：刘钊[1] 高培超 闵世平[2] 赵龙[2] 罗智德[1] 

机构地区：[1]清华大学土木工程系地球空间信息研究所,北京100084 [2]中铁二院工程集团有限责任公司,成都610031

出　　处：《国土资源遥感》2014年第3期43-47,共5页Remote Sensing for Land & Resources

摘　　要：线状模型是虚拟三维场景中的重要模型,其精细程度直接影响着虚拟场景的视觉效果。制作线状模型的常用方法分为整体模型法和分段拼接法,对于弯曲程度较大的线状模型,使用整体模型法制作时难以控制顶点坐标和纹理贴图,而采用分段拼接法又会在转弯处产生模型空隙和纹理重叠,导致模型失真。该文提出了一种针对大曲率线状实体的三维可视化方法,在几何模型制作阶段,根据路径和截面数据直接计算模型顶点坐标、法向量和索引数据,实现放样建模;在纹理贴图阶段,通过建立模型顶点与纹理图片像素的映射关系,实现模型贴图。结果表明,该算法可应用于铁路路基等大曲率线状模型的制作,具有精度可靠、人工交互少及可视化效果好的特点。The linear model is among the vital models in 3D visual scene, and its degree of fineness determines the visual effect of the scene. Common methods for constructing linear models can be divided into integral model method and spliced model method;nevertheless, for large-curvature linear models, neither method is ideal. The vertex coordinates and texture mapping are difficult to control when the integral model method is employed, and space between models and texture overlaps is inevitable when the other methods are used. A 3D visualization method for high-curvature linear entity is proposed in this paper. In the phase of geometric modeling, the vertex coordinates, normal vectors and index data are calculated for model lofting based on path and cross section. In the phase of texture mapping, textures are made by creating a mapping between vertices on the model and pixels on the photo. The results show that the algorithm can be utilized to make large -curvature linear models like railway roadbed models, and the algorithm has the merits of high accuracy, requirement of less manual work and good visual effect.

关 键 词：线状模型 三维可视化 大曲率线状实体 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置] P208[自动化与计算机技术—控制科学与工程]