利用B样条实现基于等高线的三维地形图

利用B样条曲线将等高线光栅图矢量化。根据B样条曲线插值公式得到等高线与纵横网格线的交点；根据这些史点的三维坐标值，利用线形插值分别得到纵横网格交点的高程，其均值作为网格变点的最终高程；最后利用OpenGL中的NURBS曲面接口生成三维地形图。该方法输入工作量小．数据存储量少。模拟效果较好。     

基于蜂窝状多孔固体结构的柔性物体模型研究

针对虚拟手术中柔性物体仿真的实时性和真实性要求,提出了一种基于蜂窝状多孔固体结构的力触觉再现模型。该模型利用多孔固体中的蜂窝单元构建柔性物体,将受表面压力影响的三维空间细化为蜂窝形状,基于结构力学中的标准梁理论获得蜂窝单元形变和作用力关系的解析表达式,进而实现基于物理意义的柔性物体形变仿真。模型不仅具有较好的形变计算实时性,而且通过改变蜂窝单元的弹性模量、孔穴边长和内角等参数,可以实现多种线性与非线性形变仿真。基于Falcon手控器构建的力触觉再现原型系统,验证了该模型的实时性和逼真性。     

基于参数域映射及B样条插值的三维重构方法

利用单目CCD图像进行物体表面非接触测量的核心是基于单幅图像的三维重构技术,常采用由阴影恢复形状(SFS)的方法实现三维重构。当图像分辨率较低时,通过由阴影恢复形状的方法重构的三维表面模型其分辨力较差,无法满足实际要求。为此提出了一种基于参数域映射及B样条插值的三维重构方法。采用B样条插值技术对图像进行放大处理,通过像素的参数域映射减小图像的失真及高频信息的丢失,根据放大后图像的灰度信息重构物体的三维表面模型。实验表明,基于参数域映射及B样条技术的图像插值方法很好地保护了图像的细节,利用该方法进行三维重构能够有效改善重构模型的分辨力和光顺性,为提高三维表面非接触测量精度创造了条件。     

一种最小长度约束的EMD包络拟合方法

 经典经验模态分解(Empirical Mode Decomposition,EMD)采用三次样条插值方法进行包络拟和,存在较严重的"过冲"现象。在研究该问题已有方法基础上,提出了一种基于最小长度约束的包络拟合方法,以包络曲线长度最小为目标函数,采用Lagrange求极小值法优化极值点处的导数值,然后采用分段三次Hermite函数插值方法进行包络拟合,得到平滑包络线.实验表明该方法能有效地克服三次样条插值法的"过冲"现象和分段抛物线插值法的人为弯折现象,能拟合出更平滑的包络线,使得EMD分解更准确,有效改善模态混淆问题.     

激光三维彩色人体扫描的数据重建技术

以激光三维扫描仪获得的人体数据的三维彩色数字化过程为例,讨论了删除噪声点、多传感器数据对齐、多传感器数据统一的数据预处理及人体数据重建、网格模型简化、网格平滑和网格细化的三维模型重建算法及特点,给出了人体数据处理实例,最终获得了三维人体模型.并采用直接处理彩色传感器获取的二维彩色图像实现三维彩色数据颜色边界提取方案,得到三维彩色数字化模型.     

基于曲线插值算法的巷道自动建模

通过分析Direct3D所能识别的数据文件(.X文件)格式,编程实现组织数据得到.X文件,快速自动生成三维巷道模型.并针对巷道拐弯处的圆滑问题,采用了改进的B样条曲线插值算法.实验表明,基于曲线插值算法的巷道自动建模,比建模软件建模易推广且运行速率高.     

飞行仿真中气动特性数据拟合方法研究

飞机气动特性数据是飞行仿真的基础,其精确性直接影响着飞行仿真效果.提出一种基于遗传算法的B样条曲线拟合方法,建立了飞机气动特性数据的数学模型.B样条曲线可对复杂曲线进行拟合,但数据点的参数选择影响着拟合的精度.参数选择通常有两种方法,一种是固定B样条基的节点选择参数,另一种是选定参数然后变化B样条基的节点.采用遗传算法将二者统一考虑,同时变化节点与参数使得拟合误差最小.通过气动特性数据拟合验证了该方法的有效性、精确性和鲁棒性.     

基于能量最小化控制点的B样条插值算法

针对在虚拟现实中重现物体的精确性问题,提出了采用自适应方法提取合适的特征点来插值B样条曲线的算法,实现了满足一定精度要求的数据点拟合以及控制点的反算.该算法首先通过曲率信息选取初始特征点作为型值点,然后应用线性约束能量最小化方法反算出控制点.由控制点和节点矢量拟合出B样条曲线,找出数据点与曲线偏差最大点,利用曲线复杂度指标自适应地添加新特征点,反复迭代直到满足要求的误差精度.该算法适用于逆向工程设计中重建物体和计算空间曲面控制点等问题.     

基于增益曲线拟合的单星多波束干扰源定位

单星多波束干扰源定位的公式是基于卫星的信号增益曲线建立起来的,用理论公式描述信号增益曲线存在较大偏差,针对该问题提出用拟合函数代替理论公式。建立干扰源定位方程时,用实际增益点的拟合函数代替理论公式可以减小误差。对实际增益点采用先插值后拟合的方法,插值方法分别选用线性(Linear)插值和三次样条(Spline)插值,拟合函数分别采用傅里叶函数(Fourier)和多项式函数(Polynomial)。两种插值方法分别搭配两种拟合函数产生4种拟合方法,对比分析理论公式和4种拟合方法共5种算法的性能,分别比较与实际增益的偏差以及定位误差的大小,发现拟合方法优于理论公式,且线性插值加3阶傅里叶函数拟合方法性能最优。     

基于特征点能量优化的样条插值地形算法

针对在虚拟战场环境仿真中构建大规模地形,现有的计算机硬件条件无法进行实时处理,本文提出了一种对特征点用线性约束能量最小化方法反算控制点;用三次准均匀B样条曲线插值特征点生成B样条曲线,最后利用线动成面的原理生成B样条地形曲面。其中对偏差较大点通过增加新特征点的方法进行了自动修复。通过对实际高程数据值插值地形进行实验,得出运用该方法拟合的地形图更精确、更逼真。该算法主要用于对地形中重点关注区域进行地形重构,以提高局部地区的分辨率和逼真度。     

Haptic Rendering of Visual Data for the Visually Impaired

We introduce a force-field haptic rendering method for converting visual data to haptic data. The method includes a novel framework to convert specialized 3D map models into force fields. We generate the final force fields by using structure from motion and implicit surface approximation algorithms. Visually impaired people then can learn to navigate with these maps, using off-the-shelf haptic     

Three-dimensional myocardial strain reconstruction from tagged MRI using a cylindrical B-spline model

In this paper, we present a new method for reconstructing three-dimensional (3-D) left ventricular myocardial strain from tagged magnetic resonance (MR) image data with a 3-D B-spline deformation model. The B-spline model is based on a cylindrical coordinate system that more closely fits the morphology of the myocardium than previously proposed Cartesian B-spline models and does not require explicit regularization. Our reconstruction method first fits a spatial coordinate B-spline displacement field to the tag line data. This displacement field maps each tag line point in the deformed myocardium back to its reference position (end-diastole). The spatial coordinate displacement field is then converted to material coordinates with another B-spline fit. Finally, strain is computed by analytically differentiating the material coordinate B-spline displacement field with respect to space. We tested our method with strains reconstructed from an analytically defined mathematical left ventricular deformation model and ten human imaging studies. Our results demonstrate that a quadratic cylindrical B-spline with a fixed number of control points can accurately fit a physiologically realistic range of deformations. The average 3-D reconstruction computation time is 20 seconds per time frame on a 450 MHz Sun Ultra80 workstation.     

Multimodal Interaction on Automultiscopic Content with Mobile Surface Haptics

In this work, we present interactive automultiscopic content with mobile surface haptics for multimodal interaction. Our system consists of a 40‐view automultiscopic display and a tablet supporting surface haptics in an immersive room. Animated graphics are projected onto the walls of the room. The 40‐view automultiscopic display is placed at the center of the front wall. The haptic tablet is installed at the mobile station to enable the user to interact with the tablet. The 40‐view real‐time rendering and multiplexing technology is applied by establishing virtual cameras in the convergence layout. Surface haptics rendering is synchronized with three‐dimensional (3D) objects on the display for real‐time haptic interaction. We conduct an experiment to evaluate user experiences of the proposed system. The results demonstrate that the system's multimodal interaction provides positive user experiences of immersion, control, user interface intuitiveness, and 3D effects.     

Adaptive Haptic Feedback Steering Wheel for Driving Simulators

Controlling a virtual vehicle is a sensory-motor activity with a specific rendering methodology that depends on the hardware technology and the software in use. We propose a method that computes haptic feedback for the steering wheel. It is best suited for low-cost, fixed-base driving simulators but can be ported to any driving simulator platform. The goal of our method is twofold. 1) It provides an efficient yet simple algorithm to model the steering mechanism using a quadri-polar representation. 2) This model is used to compute the haptic feedback on top of which a tunable haptic augmentation is adjusted to overcome the lack of presence and the unavoidable simulation loop latencies. This algorithm helps the driver to laterally control the virtual vehicle. We also discuss the experimental results that demonstrate the usefulness of our haptic feedback method.     

Three-dimensional statistical model for gingival contour reconstruction

Optimal gingival contours around restored teeth and implants are of critical importance for restorative success and esthetics. This paper describes a novel computer-aided methodology for building a 3-D statistical model of gingival contours from a 3-D scan dental dataset and reconstructing missing gingival contours in partially edentulous patients. The gingival boundaries were first obtained from the 3-D dental model through a discrete curvature analysis and shortest path searching algorithm. Based on the gingival shape differential characteristics, the boundaries were demarcated to construct the gingival contour of each individual tooth. Through B-spline curve approximation to each gingival contour, the control points of the B-spline curves are used as the shape vector for training the model. Statistical analysis results demonstrate that the method can give a simple but compact model that effectively capture the most important variations in arch width and shape as well as gingival morphology and position. Within this statistical model, the morphologically plausible missing contours can be inferred based on a nonlinear optimization fitting from the global similarity transformation, the model shape deformation and a Mahalanobis prior. The reconstruction performance is evaluated through large simulated experimental data and a real patient case, which demonstrates the effectiveness of this approach.     

Modeling the forces of cutting with scissors

Modeling forces applied to scissors during cutting of biological materials is useful for surgical simulation. Previous approaches to haptic display of scissor cutting are based on recording and replaying measured data. This paper presents an analytical model based on the concepts of contact mechanics and fracture mechanics to calculate forces applied to scissors during cutting of a slab of material. The model considers the process of cutting as a sequence of deformation and fracture phases. During deformation phases, forces applied to the scissors are calculated from a torque-angle response model synthesized from measurement data multiplied by a ratio that depends on the position of the cutting crack edge and the curve of the blades. Using the principle of conservation of energy, the forces of fracture are related to the fracture toughness of the material and the geometry of the blades of the scissors. The forces applied to scissors generally include high-frequency fluctuations. We show that the analytical model accurately predicts the average applied force. The cutting model is computationally efficient, so it can be used for real-time computations such as haptic rendering. Experimental results from cutting samples of paper, plastic, cloth, and chicken skin confirm the model, and the model is rendered in a haptic virtual environment.     

电镜重构病毒的三维显示与三维任意区域分割

利用冷冻电镜单颗粒三维重建方法获得分辨率1.5nm家蚕质多角体（BmCPV）;2.5nm,中蜂囊状幼虫病病毒（CSBV）;1.4nm,伊蚊C6/36细胞浓核病毒（C6/36DNV）的三维结构体数据,对三维结构数据进行了数据归并,并找到对显示细节较为敏感的向量维度,用表面显示方法对病毒体数据进行三维显示,用基于空间的方法完成了任意区域的交互分割。显示时间和信噪比与原有显示方法有了提高。病毒表面细节和轴上突起清晰可见。所分割的部分结构准确完整,并可以对所分割的部分进行量化分析。为冷冻电镜单颗粒重构提供了新的三维可视与分析方法。