基于CUDA海量空间数据实时体绘制研究

针对海量空间科学数据的精细及实时三维绘制需求,提出并实现了一种基于CUDA语言的并行化光线投射体绘制加速算法,利用传统体绘制算法中光线投射法的可并行特点和GPU中高速的纹理查询的优点,通过一个实际坐标到纹理坐标的转换函数实现了对不规则采样数据的准确采样,并完成了绘制算法的CUDA并行化改造,通过CUDA语言利用GPU强大的并行计算能力实现了对海量空间数据的实时三维光线投射绘制.     

基于CUDA的体数据可视化工具

伴随着科学技术的不断进步,计算机能力也随之提升,GPU可编程的优势以及自身的并行计算能力具有非常广阔的市场价值,正是基于此,借助光线投射等一系列运行算法,能在对数据进行交互式可视化操作后,运行相关结构算式,确保绘制效果贴合实际需求,从而提升整体可视化工具的效果,进一步创建更加优质的图像效果.本文从可视化研究背景入手,对基于CUDA的体数据可视化工具试验和单遍光线投射分析进行了集中阐释,并结合算例Phong光照计算对整体数据模型展开了讨论,旨在为相关技术研究人员提供有价值的参考建议.     

基于图形处理器的近似等值面提取算法

提出了一种新颖的等值面提取算法，该算法基于Ray-Isosurface intersection方法，通过模拟三维扫描仪的工作过程来提取等值面。算法应用GPU的并行计算能力来完成主要的计算密度大的计算过程，计算结果存储为点的形式，并通过iso-splatting的技术绘制出来。算法通过控制GPU所使用的缓冲区大小和通过模拟三维扫描仪工作原理进而避免不可见部分的等值面提取，来达到一个高质量的绘制效果和高FPS的交互环境。通过该算法和传统算法在几组数据上面的绘制质量，FPS的比较证明了算法的优越性。     

基于GPU的三维地震数据场体绘制方法

体绘制技术是计算可视化研究和应用热点之一。在对三维数据体进行形式化定义基础上,讨论光线投射算法中数据体划分,重采样计算以及图像合成的原理和方法。利用着色器进行重采样和图像合成运算,实现体绘制的GPU加速。将GPU加速的光线投射体绘制方法应用于地震数据解释,分别实现地震数据的灰度和伪彩色样式可视化,并通过转换函数,凸显出地震数据场的层位特征,克服了地震数据剖面、切片以及三维面绘制图像的局限性。     

电磁环境多层等值面融合绘制研究

针对电磁环境三维体数据中提取出的多层等值面,根据交互绘制速率与绘制质量的不同要求,提出分层伪排序融合绘制方法和改进Depth Peeling融合绘制方法。前者实现了多层等值面的近似排序,采用逐像素着色增强边界轮廓,绘制速率快,适合需要快速绘制而质量要求不高的场合;后者通过引入Slicemap技术,解决了多层等值面片元分布不均匀的问题,并利用多遍绘制融合方法,提高了绘制复杂等值面的能力,从可视化效果看,提高了绘制质量,绘制效率也得到了保证。     

基于BRDF 和GPU 并行计算的全局光照实时渲染

基于光线追踪,将屏幕图像像素分解为投射光线与场景对象交点面片辐射亮度和 纹理贴图的合成,每个面片的辐射亮度计算基于双向反射分布函数(BRDF)基的线性组合,并通 过图形处理器(GPU)处理核心并行绘制进行加速,最后与并行计算的纹理映射结果进行合成。 提出了一种基于BRDF 和GPU 并行计算的全局光照实时渲染算法,利用GPU 并行加速,在提 高绘制效率的前提下,实现动态交互材质的全局光照实时渲染。重点研究：对象表面对光线的 多次反射用BRDF 基的线性组合来表示,将非线性问题转换为线性问题,从而提高绘制效率; 利用GPU 并行加速,分别计算对象表面光辐射能量和纹理映射及其线性组合,进一步提高计算 效率满足实时绘制需求。     

一种改进的基于CUDA的纹理映射和光线投射结合的体绘制算法

针对传统的基于GPU的光线投射算法绘制效率较低的问题,利用CUDA架构的并行计算特性和对三维纹理的处理能力进行改进和优化.将体数据映射为三维纹理,利用CUDA三维数组进行存储与绑定,纹理拾取的浮点返回值利用线性滤波进行平滑.在传输函数的设计中引入中心差分梯度幅值增强对体数据边界面的绘制效果.每条光线的求交及颜色积累采用并行计算,按照由前向后进行颜色及不透明度累积.设置不透明度阈值,采用不透明度提前终止加速绘制.实验结果表明,绘制速度较传统的基于GPU算法有10％的速度提升,绘制效果也有很大的改善.     

基于简化光线投射GPU大规模地形实时可视化

地形数据处理、三角形细分等级测度和T型连接消除是基于GPU大规模地形可视化的关键问题。利用GPU简化光线投射的固定网格投射方法生成地形粗糙网格;根据地形特征和视点因素确定三角形细分等级,并匹配与存储在GPU缓存中的三角形模板,实现地形网格的自适应细分;通过改变相邻三角形细分等级的方法消除T型连接,实现基于简化光线投射GPU大规模地形的无缝绘制。实验表明该方法可以取得较高的帧速率和较好的绘制效果,实现大规模地形的实时可视化。     

基于GPU的四维医学图像动态快速体绘制

传统的三维医学图像重建技术无法满足四维医学图像动态重建的需求,而四维医学图像庞大的数据量使传统重建技术很难实现高性能实时绘制.基于以上需求,提出了一种四维医学图像动态快速体绘制方法.首先采用GPU强大的并行计算能力,提出一种基于GPU、利用CUDA技术实现的光线投射算法;然后分析了算法框架、体数据及计算结果的存储策略、...     

基于GPU的三维医学图像混合可视化系统

研究并实现了一个基于GPU的医学图像混合可视化系统，该系统采用三维纹理映射的方法实现直接体绘制，利用GPU的可编程特性完成体绘制方法中的插值后分类算法和传输函数的传递及实时修改，采用OpenGL技术实现表面的绘制，并基于场景图结构实现时表面数据的管理。面绘制和体绘制部分都采用OpenGL实现，运用OpenGL的融合机制，系统实现了面绘制和体绘制的混合显示。本系统大大提高了体绘制的速度，有效地保留了面绘制和体绘制的优势，在保证绘制速度的基础上丰富了图像信息。     

基于VTK的腹部医学图像体绘制应用研究

由于面绘制不能保留CT图像丰富的数据信息,无法良好地对于形状特征模糊的组织器官以及它们之间的层次关系进行三维显示,因此需要结合体绘制来为医生提供有力的病变分析依据。本文利用VTK开发包所提供的体绘制方法来实现腹部CT图像的三维重建,验证并比较了这几种方法各自的优缺点。实验结果表明,VTK用于CT图像的体绘制方法具有快速、效果好的特点,并可根据实际应用来选用不同的体绘制方法。     

Confetti: object-space point blending and splatting

We present Confetti, a novel point-based rendering approach based on object-space point interpolation of densely sampled surfaces. We introduce the concept of a transformation-invariant covariance matrix of a set of points which can efficiently be used to determine splat sizes in a multiresolution point hierarchy. We also analyze continuous point interpolation in object-space and we define a new class of parameterized blending kernels as well as a normalization procedure to achieve smooth blending. Furthermore, we present a hardware accelerated rendering algorithm based on texture mapping and /spl alpha/-blending as well as programmable vertex and pixel-shaders.     

Visual inspection of multivariate volume data based on multi-class noise sampling

Visualizing multivariate volume data is useful when the user wants to inspect the correlational distributions of multiple variables in a spatial field. Existing solutions commonly rely on color blending or weaving techniques to show multiple variables on a sampling point, probably causing heavy visual confusion. This paper presents an alternative solution that employs a multi-class sampling technique to generate spatially separated sampling points for multiple variables and illustrates the sampling points of each variable individually. We combine this new sampling scheme with the conventional direct volume rendering mode, iso-surface mode, and the cutting plane mode to support interactive inspection of volumetric distributions of multiple variables. The effectiveness of our approach is demonstrated with the IEEE VIS Contest 2004 Hurricane dataset and a 3D nuclear fusion simulation dataset.     

Integration of volume decompression and out-of-core iso-surface extraction from irregular volume data

Volume datasets tend to grow larger and larger as modern technology advances, thus imposing a storage constraint on most systems. One general solution to alleviate this problem is to apply volume compression on volume datasets. However, as volume rendering is often the most important reason why a volume dataset was generated in the first place, we must take into account how a volume dataset could be efficiently rendered when it is stored in a compressed form. Our previous work [21] has shown that it is possible to perform an on-the-fly direct volume rendering from irregular volume data. In this paper, we further extend that work to demonstrate that a similar integration can also be achieved on iso-surface extraction and volume decompression for irregular volume data. In particular, our work involves a dataset decomposition process, where instead of a coordinate-based decomposition used by conventional out-of-core iso-surface extraction algorithms, we choose to use a layer-based structure. Each such layer contains a collection of tetrahedra whose associated scalar values fall within a specific range, and can be compressed independently to reduce its storage requirement. The layer structure is particularly suitable for out-of-core iso-surface extraction, where the required memory exceeds the physical memory capacity of the machine that the process is running on. Furthermore, with this work, we can perform on-the-fly iso-surface extraction during decompression, and the computation only involves the layer that contains the query value, rather than the entire dataset. Experiments show that our approach can improve the performance up to ten times when compared with the results based on traditional coordinate-based approaches.     

GPU加速的光滑轮廓线绘制

轮廓线的高效提取是非真实感绘制的一个关键问题。提出了一个完全利用GPU生成光滑轮廓线的高效算法。在几何处理阶段,先根据相邻三角形的法向量与视向量的关系检测出轮廓线,然后对轮廓线进行宽度扩充,同时对轮廓线顶点设置相应的渐变因子;在像素处理阶段把渐变因子转化为相应的alpha值,通过光照生成卡通渲染,最后通过alpha混合得到光滑轮廓线。算法完全在GPU里实现,能满足实时的绘制要求。     

Fast generation and display of iso-surface wireframes

The paper presents a method for generating and displaying wireframe approximations to surfaces of constant value (or iso-surfaces). Input to the method is a data grid, a volume decomposition with each of whose vertices is associated a scalar value. During a preprocessing phase, the method constructs a threshold-independent data structure based upon the given data grid. The data structure relates the edges of an iso-surface wireframe to the edges of the data grid, for all possible threshold values. During the subsequent rendering phase, the data structure supports efficient generation and display of the iso-surface wireframe corresponding to any selected threshold value. The technique is efficient enough to form the basis for an interactive software system for visualizing iso-surfaces.     

可视化技术在发动机三维流动数值计算中的应用研究

基于MS Windows平台，应用面向对象技术，采用VisualC＋＋6．0编程，调用OpenGL图形库，我们对Marching Cubes经典算法进行了改进，实现了等值面及其体绘制、云图和流线的生成，开发了一个发动机三维瞬态数值模拟计算流场可视化系统。     

Efficient data-parallel tree-traversal for BlobTrees

The hierarchical implicit modelling paradigm, as exemplified by the BlobTree, makes it possible to support not only Boolean operations and affine transformations, but also various forms of blending and space warping. Typically, the resulting solid is converted to a boundary representation, a triangle mesh approximation, for rendering. These triangles are obtained by evaluating the corresponding implicit function (field) at the samples of a dense regular three-dimensional grid and by performing a local iso-surface extraction at each voxel. The performance bottleneck of this rendering process lies in the cost of the tree traversal (which typically must be executed hundreds of millions of times) and in the cost of applying the inverses of the space transformations associated with some of the nodes of the tree to the grid samples.Tree pruning is commonly used to reduce the number of samples for which the field value must be computed. Here, we propose a complementary strategy, which reduces the costs of both the traversal and of applying the inverses of the blending and warping transformations that are associated with each evaluation.Without blending or warping, a BlobTree can be reduced to a CSG tree only containing Boolean nodes and affine transformations, which can be reordered to increase memory coherence. Furthermore, the cumulative effects of the affine transformations can be precomputed via matrix multiplication. We propose extensions of these techniques from CSG trees to the fully general BlobTrees. These extensions are based on tree reordering, bottom-up traversal, and caching of the combined matrix for uninterrupted runs of affine transformations in the BlobTree.We show that these new techniques result in an order of magnitude performance improvement for rendering large BlobTrees on modern Single Program Multiple Data (SPMD) devices.     

ALPHA融合在无缝投影中的应用

由于大屏幕投影的快速发展和广泛应用,大屏幕投影中的无缝融合技术也日益成为研究和应用的热点.在对alpha融合进行探讨和研究基础上,进行了一定的改进,提出并实现了一种基于alpha融合的新线性算法.算法在alpha通道的基础上,提出了对图像投影后重叠的边缘部分采取线性变换,降低了图像的透明度,达到无缝融合的目的.     

Projection Mapping on Arbitrary Cubic Cell Complexes

This work presents a new representation used as a rendering primitive of surfaces. Our representation is defined by an arbitrary cubic cell complex: a projection‐based parameterization domain for surfaces where geometry and appearance information are stored as tile textures. This representation is used by our ray casting rendering algorithm called projection mapping, which can be used for rendering geometry and appearance details of surfaces from arbitrary viewpoints. The projection mapping algorithm uses a fragment shader based on linear and binary searches of the relief mapping algorithm. Instead of traditionally rendering the surface, only front faces of our rendering primitive (our arbitrary cubic cell complex) are drawn, and geometry and appearance details of the surface are rendered back by using projection mapping. Alternatively, another method is proposed for mapping appearance information on complex surfaces using our arbitrary cubic cell complexes. In this case, instead of reconstructing the geometry as in projection mapping, the original mesh of a surface is directly passed to the rendering algorithm. This algorithm is applied in the texture mapping of cultural heritage sculptures.