时空间混合并行框架及其在并行计算实验教学中的应用

结合处于研究前沿的时间并行理论和传统的空间并行技术,设计开发了一个时空混合并行计算框架,用于并行计算和数值模拟相关课程的实验教学。该框架提供了基础的时空间并行功能和部分常用的数值求解器,又支持在更高层面完成对问题的建模和分解,有助于学生全面理解和掌握科学计算和工程模拟的相关知识,同时便于积累教学成果,探索和培育新课题。     

Altemative Monte Carlo Approach for General Global Illumination

An alternative Monte Carlo strategy for the computation of global illumination problem was presented.The proposed approach provided a new and optimal way for solving Monte Carlo global illumination based on the zero variance importance sampling procedure. A new importance driven Monte Carlo global illumination algorithm in the framework of the new computing scheme was developed and implemented. Results, which were obtained by rendering test scenes, show that this new framework and the newly derived algorithm are effective and promising.     

Registration Method for CT-MR Image Based on Mutual Information

Medical image registration is the process of finding ageometric transformation between two image-based coordi-nate systems that have the same patient-based coordinates ,i .e .the same anatomic locations .The registered multi-mo-dality images lead to impro…     

Dynamic Distribution Model with Prime Granularity for Parallel Computing

Dynamic distribution model is one of the best schemes for parallel volume rendering. However, in homogeneous cluster system, since the granularity is traditionally identical, all processors communicate almost simultaneously and computation load may lose balance. Due to problems above, a dynamic distribution model with prime granularity for parallel computing is presented. Granularities of each processor are relatively prime, and related theories are introduced, A high parallel performance can be achieved by minimizing network competition and using a load balancing strategy that ensures all processors finish almost simultaneously, Based on Master-Slave-Gleaner （ MSG） scheme, the parallel Splatting Algorithm for volume rendering is used to test the model on IBM Cluster 1350 system. The experimental results show that the model can bring a considerable improvement in performance, including computation efficiency, total execution time, speed, and load balancing.     

Alternative Monte Carlo Approach for General Global Illumination

Ithasbeenwellknownthatglobalilluminationsimu latesthelighttransportamongallthesurfacesandthelightsourcesinavirtualsceneconsideringdiffusereflection/refrac tionandspecularreflection/refraction .Duetosimplicity,generalityanddimensionalindependence,MonteCarlomethodorrandomwalkmethodisoneofthemostimportantmethodsforsolvingthedifficultglobalillumination .Inpartic ular,MonteCarloisthelastresortwhenallotheranalyticalornumericalmethodsfail[1 ] .　　Lighttransportormultiplelightinteractionsamongthesu…     

中国水墨画效果的计算机仿真

基于长期的绘画实践 ,对中国水墨画的各组成要素的材料特性及相互间作用关系进行了分析 ,提出分解水墨画效果的研究方法 ,给出了仿真模型体系 .借鉴粒子系统的基本概念 ,引入流体仿真技术及层对象混合算法 ,实现了中国水墨画的典型艺术效果     

Reconstruction of Novel Viewpoint Image Using GRNN

Theaimofthecomputergraphicsistoachieverea listicimage .Traditionalmethods ,suchasradiosity ,hav ingconstructedthemodelofthescene,produceimagesbyilluminationalgorithm .Conventional graphicsfacegreatdifficultiesintwoaspects :over complicatedcom putingandunrealisticimages.Theimage basedmodelingandrendering(IBMR)techniquesareapowerfulalterna tivetothetraditionalgeometry basedtechniquesforim agesynthesisandarewidelystudiedinrecentyears .Im age basedrendering(IBR)techniquesdirectlyrenderthenovelv…     

利用PVM实现整体光照的并行计算

对利用网络资源解决复杂且耗时的计算问题做了尝试 .选择典型的粒子跟踪整体光照计算问题作为研究对象 ,提出一种改进的且充分发掘其内在相关特性的密度估计算法 .在以太网连接的多台微机上 ,以PVM (并行虚拟机 )机制实现了该算法的并行计算 .通过对各项运算性能指标的测试与分析 ,结果显示可获得良好的加速比 ,并且PVM在分布式网络并行计算上将有很好的应用前景     

Reconstruction of Novel Viewpoint lmage Using GRNN

A neural-statistical approach to the reconstruction of novel viewpoint image general regression neural networks(GRNN) is presented.Different color value will be obtained by watching the same surface point of an abject from different viewpoints due to specular reflection,and the difference is related to the position of viewpoint.The relationship between the position of vewpoint and the color of image is non-linear,neural network is introduced to make curve fitting,where the inputs of neural network are only a few calibrated images with obvious specular reflection.By training the neural network,network model is obtained.By inputing an arbitrary virtual viewpoint to the model,the image of the virtual viewpoint can be computed.By using the method presented here,novel viewpoint image with photo-realistic property can be obtained,especially images with obvious specular reflection can accurately be generated.The method is an image-based rendering method,geometric model of the scene and position of lighting are not needed.     

Non-Rigid Image Registration Algorithm Based on B-Splines Approximation

An intensity-based non-rigid registration algorithm is discussed, which uses Gaussian smoothing to constrain the transformation to be smooth, and thus preserves the topology of images. In view of the insufficiency of the uniform Gaussian filtering of the deformation field, an automatic and accurate non-rigid image registration method based on B-splines approximation is proposed. The regularization strategy is adopted by using multi-level B-splines approximation to regularize the displacement fields in a coarse-to-fine manner. Moreover, it assigns the different weights to the estimated displacements according to their reliabilities. In this way, the level of regularity can be adapted locally. Experiments were performed on both synthetic and real medical images of brain, and the results show that the proposed method improves the registration accuracy and robustness.