一种基于GPU实现的自适应八叉树纹理绘画算法

当前,虽然基于二维图像映射定义的传统二维纹理已得到广泛应用。但是它有很多局限性。这是因为很多三维模型在纹理空间中进行参数化是非常困难的,例如隐式表面、细分表面和高密度或高细节的多边形网格。基于八叉树纹理定义,提出了一种新型的自适应八叉树纹理绘画算法。和传统的八叉树纹理映射算法相比,不但占用更少的存储空间,而且实现了基于GPU的纹理查询,有更快的查找速度。     

由多视图近似重构三维物体形态的方法及应用

本文介绍由多视图近似重构三维物体的基本方法，文中采用三维实体表达的八叉树方法，给出了表达视锥的基本数据模型及构造视锥八叉树的基本方法。本文所提供的重构方法已应用于高坝泄洪时雾化区重构的研究。     

对基于二次误差的模型简化方法的改进

模型简化在虚拟现实、模拟仿真等方面有着广泛的应用,但算法复杂、时间复杂度高一直是模型简化的主要问题。为了提高算法速度,沿用Garland提出的使用二次误差矩阵来简化三角面片模型的方法,使用八叉树空间分割的方法加速模型简化算法中用于进行边收缩的候选最近点对的排序过程,从而提高算法的速度。     

曲面激光密集测量三维数据的三角片逼近方法

本文以激光－机器视觉测量方法得到的曲面数据云为基础，探讨了曲面密集三维散乱点群数据的几何建模方法，根据激光测量方法和三维点群分布的特点，建立了恰当的数据结构在计算机中表示散乱点群。由八叉树空间分割原理对密集散乱点群进行空间分割，建立八叉树拓扑关系。由八叉树空间分割原理对密集散乱点群进行空间分割，建立八叉树拓扑关系，显著加快了任意点的搜寻速度。采用万有引力定律计算三角片顶点坐标，由此实现散乱点数据的     

基于八叉树编码的点云数据精简方法

针对逆向工程中的点云数据预处理,分析了现有数据精简方法的不足,提出了基于八叉树编码的均匀精简方法。应用八叉树编码法划分点云邻域空间为多个指定边长的子立方体,保留每个子立方体中距中心点最近的点,实现从空间整体角度对点云的精简。对涡轮叶片测量数据进行了精简测试,证明了该算法的有效性和实用性。     

基于八叉树邻域分析的光线跟踪加速算法

八叉树是加速光线跟踪常用的层次划分结构,为加快八叉树跟踪光线的过程,论文研究了运用八叉树邻域分析提高光线与八叉树节点之间的碰撞检测速度的方法,提出了一种结构简单、计算效率更高的八叉树节点的邻域分析算法。运用该算法可由现碰撞节点快速计算出下一碰撞节点,避免了采用大量递归搜索计算,从而提高了图像的渲染速度。实验结果表明,使用论文提出的邻域分析进行碰撞检测,效率比传统算法提高了3倍以上,大大提高了光线跟踪的速度。     

抽象物体的2n-叉树表示及其在高维布局问题中的应用

针对高维几何布局问题中布局物体的建模问题,提出2^n－叉树方法,在一般维度下系统地给出了抽象物体的叉树表示法的构造。物体的2^n0－叉树表示是物体八叉树表示理论的向高维扩展,同时也包容八叉树方法向低 维问题的推广。     

货物空腔背景下基于遗传算法的集装箱装载设计优化研究

集装箱装载问题以假设货物为规则实体进行研究,方便了相关求解算法的改进,但是忽略货物本身空腔的存在,已经开始阻碍集装箱装载率的进一步提高。对于现有文献很少研究的货物空腔背景下的集装箱装载问题,设计一种基于遗传算法的空间优化模型。模型通过空间预处理,将货物按照一定规则嵌套填充,使不可用或小型货物空腔叠加成较大的可用空间;定义一种四叉树空间结构来表达空间的分解;提出并运用四空间分割法来设计装载策略;引入自适应遗传算法,进行模型寻优。最后案例分析表明,该模型可以大大减弱空腔对整体装载率的降低作用,在实际应用时具有一定的参考价值。     

虚拟手术中实时碰撞检测技术研究

碰撞检测是虚拟手术的关键技术,为提高检测速度,满足系统实时性的要求,提出空间剖分和层次包围盒相结合的方法。使用八叉树表示法对虚拟场景进行空间剖分,在叶节点构建层次包围盒。进行碰撞检测时属于不同八叉树节点的几何元素不会相交,否则使用层次包围盒算法继续进行检测,对于有可能相交的几何元素再进行精确相交检测。     

实体数控仿真刀具扫掠体的构造

实体数控仿真具有重要意义，刀具扫掠体构造是实体数控铣削仿真的重要环节．分析了刃具轮廓、刀具运动形式和加工指令对构造刀具扫掠体的影响．基于ACIS平台研究了刀具扫掠体的构造，提出了动态构造刀具扫掠体算法，给出了该算法的流程图．并将其应用到正在研制和开发的实体数控铣削仿真系统中，给出了2．5D实体数控铣削仿真的应用实例。     

基于八叉树的复杂产品模型实时绘制技术

给出基于OpenSceneGraph场景模型的信息提取方法,利用八叉树对场景模型进行分割和视锥体剔除,有效提高了实时绘制的效率,尤其是对浏览场景细节时的绘制效率提高最为明显。采用基于分页技术的Pagelod方法,实现模型的动态调度,以减少I/O的负载,满足在有限硬件条件下虚拟场景中复杂产品模型的实时绘制要求。     

光学照像测量及数据压缩方法研究

光学照像测量技术的发展大大提高了逆向工程技术中的数据测量速度，使快速方便地获取包含被测物体更多细节的大量数据成为可能，但庞大的点云数据影响了模型重建的效率和质量。如何压缩散乱数据点已成为逆向工程中一个重要的研究课题，本文在介绍光学照像原理的基础上，分析常用数据压缩方法以及一种以八叉树原理为基础的的数据压缩算法。     

Nonlinear signal-processing model for scalar diffraction in optical recording

A nonlinear signal-processing model is derived for the optical recording channel based on scalar diffraction theory. In this model, the signal waveform is written in closed form as an explicit function of the channel bits that are stored on an optical disk, thereby comprising both linear and nonlinear terms. Its explicit dependence on the channel bits makes this model well suited for signal-processing purposes. With the model it is also convenient to assess the importance of nonlinear contributions to the signal waveform. The model is applied for one-dimensional optical storage as well as for two-dimensional (2D) optical storage in which bits are arranged on a 2D hexagonal lattice. Signal folding is addressed as a typical nonlinear issue in 2D optical storage and can be eliminated by recording of pit marks of sizes considerably smaller than the size of the hexagonal bit cell. Further simplifications of the model with only a limited number of channel parameters are also derived.     

Compaction Properties of Powders: The Relationship Between Compression Cycle Hysteresis Areas and Maximally Applied Punch Pressures

The consolidation behaviors of various pharmaceutical solids were characterized by investigating the relationship between the calculated hysteresis areas and the maximally applied punch pressures. An Instron universal testing apparatus and an instrumented die were used to generate compression cycle profiles at various maximally applied punch pressures for the materials studied. Based on the profiles obtained, hysteresis areas were calculated for the materials studied as a function of maximally applied punch pressures. Furthermore, model profiles describing the plastic and brittle fracture processes were utilized to derive mathematical relationships between the calculated hysteresis cycle areas and the maximally applied punch pressures. The mathematical relationships derived indicate that a linear relationship between hysteresis areas and maximally applied punch pressures exists for plastic materials, whereas for brittle materials the hysteresis areas are related to the square of the maximally applied punch pressures. Experimental data obtained support the mathematical relationships derived. The goodness of fit to the models derived is used to rank order the consolidation mechanism of various drugs and pharmaceutical excipients.     

Helmholtz Resonances in Photoacoustic Experiment with Laser-Sintered Polyamide Including Thermal Memory of Samples

A minimum volume photoacoustic (PA) cell, employed as an acoustic resonator, is presented. A model of the PA cell transfer function, as a combination of two Helmholtz resonators, is derived. Frequency PA response is described as the product of theoretically derived pressure variation and the transfer function of the PA cell. The derived model is applied to the frequency PA measurements, performed on laser-sintered polyamide. It is shown that the derived model explains the occurrence of resonant peaks in the high-frequency domain (\({>}\)10 kHz), in both amplitude and phase measurements, obtained by a gas-microphone minimal volume PA cell. The implementation of the model in the growing possibilities of sample characterization using gas-microphone PA cell is discussed.     

Mixed- and general-order kinetics applied to selected thermoluminescence glow curves

Mixed-order (MO) and general-order (GO) kinetics expressions are applied to experimental glow curves of CaSO4:Ce, LiNaSO4:Eu, BaF2:Ce and SrF2:Er. The purpose is to compare the activation energies derived from the two models to investigate the correlation between the order of kinetics, b, and the parameter alpha of the MO model for real systems and to explore the validity of the correlation between b and alpha derived from the analysis of synthetic glow peaks or experimentally isolated single peak for complex glow curves. The two alternative routes resulted in clean fits with very close values of the sum of squared residuals. The general conclusions are: (1) the activation energies derived from the MO model are slightly higher than the ones derived from the GO model, but the difference appears to be insignificant, (2) the correlation between b and alpha is not smooth and the scatter in the b values for a given alpha is within the theoretically expected spread in the b value and (3) the MO expression is capable of evaluating the shape parameters as accurately as the GO expression with the advantage that it has a physical basis contrary to the purely empirical GO model.     

Employment of tandem mass spectrometry for the accurate and specific identification of oligosaccharide structures

A method is described for the rapid identification of oligosaccharides employing a library of tandem MS spectra. Identification is aided by software that compares the sample tandem MS to those in the library. The method incorporates quadrupole time-of-flight mass spectrometry along with an annotated oligosaccharide (OS) structure library and the MassHunter Personal Compound Database and Library (PCDL) software. With an automated spectra search, OS structures in different samples are readily identified. This method is shown to be useful in the study of milk oligosaccharides but can be readily applied to oligosaccharide pools in other biological tissues.     

A static fatigue model for the durability of glass fibre reinforced cement

This paper presents a model for predicting service lives for glass-fibre reinforced cement (grc) components using hot-water accelerated ageing. It improves on previous models, being derived from consideration of a specific proposed micro-mechanical strength loss mechanism based on static fatigue principles and can be applied from time = 0. The model fitted well to all available strength vs. time data pertaining to various grc formulations. The activation energies thus derived for the strength loss process (80–90 kJ mol^–1) were consistent with those derived previously and those proposed for general glass dissolution mechanisms. Updated acceleration factors for predicting service lives for grc are advanced. The model was also applied to grc made with modified cement matrices. For metakaolin modified matrices, the activation energy appeared similar to that for OPC-grc, thus the use of similar acceleration factors appears justified. There is some evidence that calcium sulphoaluminate modified grc degrades according to a different activation energy. More data are required for modified matrix grcs if the model is to be applied thereto with confidence.     

Primal and mixed upwind finite element approximations of control advection-diffusion problems

Control advection-diffusion problems are formulated via variational inequalities and effective upwind finite element approximations are studied. The method of local subdifferentials is applied to model and dualize control constraints, as well as to produce global primal and mixed variational formulations. Upwind finite element schemes are derived, satisfying the discrete maximum principle and the conservation of mass law. The numerical resolution methods used are iterative algorithms of the Uzawa type, which are formulated and analyzed. Some numerical experiments are presented for a model discrete problem.     

Grain size statistics

A method for the computation of spatial grain size distributions from intercept data based on a tetrakaidecahedron grain model is developed. The necessary inverse matrix is presented. The method is applied to a range of metallic and ceramic specimens. The derived distributions are analysed to show that they are not necessarily log-normal. Statistical techniques are applied to determine the minimum sample sizes and these are shown to increase as the distributions become more dispersed. The constant relating the average grain size to the average intercept length is also shown to be sensitive to the grain size distribution.