 上传我的文档
 下载
 收藏
该文档贡献者很忙，什么也没留下。
 下载此文档
正在努力加载中...
基于视频的火焰检测算法研究
下载积分：1000
内容提示：基于视频的火焰检测算法研究
文档格式：PDF|
浏览次数：0|
上传日期： 08:41:04|
文档星级：
全文阅读已结束，如果下载本文需要使用
 1000 积分
下载此文档
该用户还上传了这些文档
基于视频的火焰检测算法研究
官方公共微信上传用户：mlpyohvlbn资料价格：5财富值&&『』文档下载 ：『』&&『』学位专业：&关 键 词 ：&&&&&&&权力声明：若本站收录的文献无意侵犯了您的著作版权，请点击。摘要:（摘要内容经过系统自动伪原创处理以避免复制，下载原文正常，内容请直接查看目录。）跟着盘算机硬件的成长和互联网技巧的普及，和片子殊效、电子游戏的飞速成长，年夜众对画面精巧度和对画面真切度的赓续寻求，急切须要进步盘算灵活画的模仿速度，天然景不雅的模仿就是个中很主要的一项模仿。最近几年来，天然景不雅的模仿成为盘算机图形学的热门之一，个中火焰等气体的模仿吸引了浩瀚图形学研讨者的普遍存眷。它们可以运用于片子殊效、视频游戏和告白等行业，发生了伟大的社会和经济效益。跟着盘算机技巧的赓续提高和流体动力学的赓续成长，火焰模仿技巧也在赓续的成长和提高当中。传统的基于粒子生成的火焰，年夜致能相符火焰的外型，普通情形下速度很快，但具有很年夜的随机性，粒子多的时刻速度较慢。纹理技巧办法生成的火焰，年夜致能相符火焰的外型，盘算量较小，盘算进程绝对简略，根本能相符熄灭的场景，但存在人工绘制的陈迹。是以火焰模仿算法研讨任务具有主要的实际意义。本文在概述了火焰衬着的各类研讨办法和最近几年来国际内在基于物理的火焰衬着范畴研讨趋向的基本上，采取一种基于物理模子的火焰及时衬着办法。本文的重要研讨目的是提出并完成基于物理模子算法的改良，使得基于物理模子算法模仿速度更快，人机的交互性更强，生成的后果图更真切，以推进火焰模仿技巧疾速向前成长。本文的研讨任务重要包含以下三个方面：1。本文剖析了火焰模仿技巧的研讨配景及意义，总结了现有各类火焰模仿算法，并提涌现阶段火焰模仿算法的成长偏向。研讨并总结现有算法中基于粒子体系、纹理体系、物理模子的优缺陷，为更进一步进步火焰模仿算法供给实际支撑。2。本文胪陈了无限差分法、无限元法、无限容积法的界说，并比拟了这三种数值解法的优缺陷，决议运用无限差分法求解Navier一Stokes方程，运用Helmholtz一Hodge分化公式将偏微分方程分化，将偏微分方程分化成若干求解便利的子方程，然后再求解各子方程，采取半拉格朗日法求解平流项，应用粒子体系特色修正平流方程，应用隐式迭代办法停止求解分散方程，可以以随意率性步长来停止火焰模仿，最初将各子方程求解的解集联合在一路而获得终究的方程解。3。本文起首引见了GPGPU的概述，随之又引见了CUDA的概述，在盘算进程中因为盘算机硬件GPU壮大的并行盘算才能，又因为子方程具有很高的盘算密度，且子方程具有雷同且彼此自力的运算步调，所以很合适用GPU做并行盘算来停止加快模仿。最初应用Jacobi迭代法求解线性方程组，应用光线投射算法对火焰停止衬着，并对光线投射算法停止并行盘算，最初生成了真切的火焰图象。试验成果注解：该办法完成简略，衬着速度快，显示的后果真实。Abstract:Rapid growth along with the computer hardware and the growth of Internet technology popularization, and movie special effects, video games, masses of the picture is exquisite and picture of the real degree of continuously seeking urgently needed progress calculation flexible painting imitation speed and natural landscape of imitation is medium is mainly an imitation. In recent years, the imitation of natural scenery has become one of the hottest computer graphics, the flame and other gases to attract a vast number of computer graphics researchers generally pay close attention to. They can be used in the industry, such as film and video games, advertising and other industries, which have great social and economic benefits. Along with the ceaseless development of the computer technology and continuously improve and fluid dynamics, flame imitation skills also in ceaseless grow and improve. Traditional particle based flame, the eve of the flame can be consistent with the shape of the flame, the general situation is very fast, but with a very large number of random, more than the time of the particle. Texturing technique way of generating the flame, the eve of the can match flame appearance, small amount of calculation, the calculating procedure relatively simple, basic can match out of the scene, but there are traces of artificial rendering. It is the main practical significance to study the task of the flame imitation algorithm. In this paper an overview of the flame lined with all kinds of research methods and in recent years international internal flame based on physical lined with category research trend basically, take a flame based on physical model of timely lining the way. The main purpose of the study is to put forward and implement improvement of physical model algorithm based on, the physical model algorithm based on imitation faster, man-machine interaction more generated consequences figure more real, in order to promote the flame imitation skills to quickly grow up and move on. The research task of this paper includes the following three aspects: 1. This paper analyzes the research background and significance of the flame imitation skills, summarizes the existing types of flame imitation algorithm, and put forward the development phase of the flame to imitate the growth of the algorithm. Research and summarize the existing algorithms based on particle system, texture system, physical model of the advantages and disadvantages, for the further progress of the flame mimic algorithm to provide the actual support. 2. This paper detailed the infinite difference method, infinite element method, infinite volume method is defined, and the advantages and disadvantages of these three kinds of numerical solution, decided to use the infinite difference method for solving the Navier Stokes equation, using a Helmholtz Hodge differentiation formula the partial differential equations of differentiation, the partial differential equations into a number of convenience for differentiation the sub equation, and then solve the equations, adopt semi Lagrange method for solving the advection, using particle system features modified advection equation, using implicit iterative approach to solve the dispersion equation, can be arbitrarily step to stop the flame imitation, will initially solve the sub equation solution set together and get after all, the solution of equation. 3. Overview of this paper first introduces the overview of GPGPU, subsequently also introduced CUDA, in the calculation process because computer hardware GPU and expansion of the parallel computing ability and because equation has very high computing density and sub equation is identical and independent operation pace. So it is suitable using GPU to do parallel computing to stop accelerating imitation. Initial application of Jacobi iterative method for solving linear equations, the use of ray casting algorithm to stop the flame lining, and the ray casting algorithm to stop parallel computing, initially generated a real flame image. Test results: the method is simple, with a fast speed, the results show that the real.目录:摘要6-8ABSTRACT8-9第1章 绪论10-17&&&&1.1 研究背景与意义10-11&&&&&&&&1.1.1 电子游戏10-11&&&&&&&&1.1.2 电影特效11&&&&&&&&1.1.3 互动投影11&&&&1.2 研究方法分类11-14&&&&&&&&1.2.1 粒子系统12&&&&&&&&1.2.2 纹理技术12&&&&&&&&1.2.3 细胞自动机12-13&&&&&&&&1.2.4 分形几何13&&&&&&&&1.2.5 基于物理模型的方法13-14&&&&1.3 研究方向14-15&&&&&&&&1.3.1 细节加强策略14-15&&&&&&&&1.3.2 流体加速策略15&&&&&&&&1.3.3 流体控制策略15&&&&1.4 主要研究内容15-16&&&&1.5 本文的组织结构16-17第2章 物理模型的三维火焰技术概述17-22&&&&2.1 基于物理的火焰模拟17-22&&&&&&&&2.1.1 流体的物理模型17-18&&&&&&&&2.1.2 拉格朗日法和欧拉法18&&&&&&&&2.1.3 常用的数值解法18-20&&&&&&&&2.1.4 方程的简化处理20-22第3章 GPU 加速技术概述22-32&&&&3.1 可编程图形流水线22-24&&&&3.2 GPGPU 介绍24-26&&&&3.3 CUDA 概述26-32&&&&&&&&3.3.1 CUDA C 语言27-28&&&&&&&&3.3.2 nvcc 编译器28&&&&&&&&3.3.3 CUDA 编程模式28-29&&&&&&&&3.3.4 CUDA 存储模式29-32第4章 Navie-Stokes 方程求解32-41&&&&4.1 有限差分法32-35&&&&&&&&4.1.1 计算区域的离散32-34&&&&&&&&4.1.2 控制方程的离散34-35&&&&4.2 求解 Navie-Stokes 方程35-38&&&&&&&&4.2.1 求解方法35&&&&&&&&4.2.2 利用粒子系统修改对流方程35-36&&&&&&&&4.2.3 采用隐式方法求解扩散方程36-37&&&&&&&&4.2.4 采用向前差分方法求解外力方程37-38&&&&&&&&4.2.5 采用隐式方法求解泊松方程38&&&&&&&&4.2.6 采用中心差分格式求解修正方程38&&&&4.3 求解线性方程组38-40&&&&4.4 实验结果分析40-41第5章 火焰的绘制41-46&&&&5.1 可视化渲染41-42&&&&&&&&5.1.1 密度场41&&&&&&&&5.1.2 光线投射41-42&&&&5.2 绘制的 CUDA 实现42-44&&&&&&&&5.2.1 算法流程42-44&&&&5.3 细节增强44-45&&&&&&&&5.3.1 漩涡控制44&&&&&&&&5.3.2 边界处理44-45&&&&5.4 实验结果45-46第6章 总结与展望46-47&&&&6.1 回顾总结46&&&&6.2 未来工作46-47参考文献47-49攻读硕士期间发表的文章49-50致谢50分享到：相关文献|相关硕士文献推荐
频道总排行
频道本月排行种基于视频多特征融合的火焰识别算法_图文_百度文库
两大类热门资源免费畅读
续费一年阅读会员，立省24元！
种基于视频多特征融合的火焰识别算法
阅读已结束，下载文档到电脑
想免费下载更多文档？
定制HR最喜欢的简历
下载文档到电脑，方便使用
还剩3页未读，继续阅读
定制HR最喜欢的简历
你可能喜欢