基于元胞自动机的自然云模拟
|
摘要
自然现象的模拟是近十几年来计算机图形学的一个重要研究领域,自然景物的渲染是图形学技术研究中的一个重点和难点之一,如何构造结构复杂、形态变化的物体是计算机图形学的一个新课题,而其中云的模拟已经成为一个研究的热点。鉴于传统的集合建模方法难以有效的对云这种形状毫无规则并且动态变化极其复杂的物体进行描述等问题,本文选用元胞自动机模型为基础,通过结合流体力学中关于描述像云这种无规则物体等方面的介绍,在云的建模方法、动态模拟以及光照模型三个方面进行了改进研究。本文的研究内容主要包括:
1.基于元胞自动机的云的建模
通过对元胞自动机模型的研究,利用其简单、高效等的特点,结合云的物理现象,将元胞自动机应用于自然云的建模中,初步实现了自然云的模型建立,一定程度上提高了建模的速度。提出了一种基于元胞自动机的云的状态转变规则,该规则以真实世界中云的特征转变为基础,对云的产生加以控制,使模型的建立符合自然云的状态转变规律。
2.结合数学物理方程研究了动态的自然云实现方法。
在对流体力学中关于描述像云这种无规则物体等方面有了一定的研究的基础上,提取出影响云的运动的几种自然因素以及它们的影响方式,提出了利用元胞自动机模型对云的动态现象进行模拟的方法。该方法能够结合自然云运动的物理原理,使用元胞自动机的状态转变规则进行实现,有助于提高模拟的真实感。在真实世界中,云的高度会对风的大小产生影响,从而影响其被动的运动;而外界环境中的浓度,温度以及产生的对流扩散等都会使其发生内部的扰动变化。在数学物理方程中,对这些变化和影响都有着较为成熟的研究。而结合已有的数学物理方程,构建出基于元胞自动机模型的状态转变规则,使得云的动态模拟即具有真实性,又不失速度。
3.对自然云的光照渲染进行了深入的研究
在自然界中,光线在介质中的传播通常会产生反射、折射、衍射、散射等现象,在云中的传播也不例外。将云看成一个个小粒子,各个粒子所接受到的光线以及由此产生的反射、散射等现象会产生相互影响,并进一步影响所接受到的光线的强度,因此,光线在云中的传播是一个复杂而反复的过程。在光线传播的简化过程中,产生了多重散射和单向散射两种理论。在本文的研究过程中,在不失真实性的前提下,将其进一步简化,采用前向为主的多重散射和单向散射相结合的方法进行光线的渲染。在光线由光源传播到云的过程中,采用前向为主的多重散射模型,即真实的表现出了光线在云中的各种传播路径,又由于对其产生的极小影响的光线进行了弱化忽略,提高了计算速度;在计算光线由云团传播到视角的过程中,采用单向散射技术,满足了真实性和速度两方面的需求。
本文中采用元胞自动机模型进行模拟,结合云的各种物理特性,从而得到了动态的,真实感强的自然云的模拟,在真实感和速度上均取得了比较满意的结果。
Over the past decade, one of the important fields in the study of Computer Graphics is the imitation of natural phenomenon, rendering of natural scene is a hot topic in the research area of graphics. How to construct complex structures, and morphological changes of the object is a new topic in computer graphics with the imitation of clouds as its hot topic. The traditional set of modeling methods can not effectively describe shapes of the clouds that have no rules, and can dynamic changes. In this paper we bases on cellular automata model and the description of clouds-like objects in hydrodynamics to further study the cloud modeling, dynamic simulation, and the illumination model. The following aspects are studied in this paper:
1. model of clouds on the basis of cellular automation.
The cellular automation is simple and high efficient. The application of cellular automation to the setting of clouds model can basically increase the speed. This paper proposed state transition rules of clouds, based on cellular automata. The rules base on the characteristics of cloud changes in the real world, and control the speed of the formation of clouds, so the model can meet the natural cloud’s state transition laws.
2. Combine with Mathematics Physics Equation to realize the natural cloud.
After studying the fluid mechanics, which describes cloud-like objects that has no rules, we extracted some natural factors that affect the movement of clouds and their affecting ways. We proposed the cloud simulation method of using of cellular automaton model to stimulate the dynamic phenomena of the cloud. This method can be combined with the natural physics of cloud movement and the using of cellular automata rules helps to improve simulation realism. In the real world, cloud height will affect the size of the wind which will affect the passive movement of the clouds; and the external environment such as the concentration, temperature and convection-diffusion, will bring about internal disturbances. These changes have been studied in mathematical physics equations. This paper uses the equations of mathematical physics to build state transition rules based on cellular automata. So the dynamic simulation of the cloud is real and speedy.
3. A deep study on the illumination rendering of natural clouds.
In nature, light propagation will reflect, refract, diffract, scatter in the medium and it is the same when it’s in the cloud. The cloud is made up of small particles, and all particles receive light and they reflect, scatter. These phenomena will have mutual influences, and further affect the strength of their received light. So the spread of light in the clouds is a complex and iterative process. There are multiple scattering theories and single scattering theories in the process of light propagation. This paper further simplifies the process of light propagation without losing the authenticity. We mainly uses the multiple forward scattering and single scattering to render the light. In the process of the light propagates to clouds, we uses the multiple scattering models, which showed various communication paths, and because of the light was weakened and ignored, this model can improve the speed; and when we calculates the light transmits from the clouds to view, we use the one-way technology, which can meet authenticity and speed.
This paper uses cellular automata model and various physical properties of clouds, to achieve a dynamic, and realistic simulation of natural cloud, which is real and speedy.
1.基于元胞自动机的云的建模
通过对元胞自动机模型的研究,利用其简单、高效等的特点,结合云的物理现象,将元胞自动机应用于自然云的建模中,初步实现了自然云的模型建立,一定程度上提高了建模的速度。提出了一种基于元胞自动机的云的状态转变规则,该规则以真实世界中云的特征转变为基础,对云的产生加以控制,使模型的建立符合自然云的状态转变规律。
2.结合数学物理方程研究了动态的自然云实现方法。
在对流体力学中关于描述像云这种无规则物体等方面有了一定的研究的基础上,提取出影响云的运动的几种自然因素以及它们的影响方式,提出了利用元胞自动机模型对云的动态现象进行模拟的方法。该方法能够结合自然云运动的物理原理,使用元胞自动机的状态转变规则进行实现,有助于提高模拟的真实感。在真实世界中,云的高度会对风的大小产生影响,从而影响其被动的运动;而外界环境中的浓度,温度以及产生的对流扩散等都会使其发生内部的扰动变化。在数学物理方程中,对这些变化和影响都有着较为成熟的研究。而结合已有的数学物理方程,构建出基于元胞自动机模型的状态转变规则,使得云的动态模拟即具有真实性,又不失速度。
3.对自然云的光照渲染进行了深入的研究
在自然界中,光线在介质中的传播通常会产生反射、折射、衍射、散射等现象,在云中的传播也不例外。将云看成一个个小粒子,各个粒子所接受到的光线以及由此产生的反射、散射等现象会产生相互影响,并进一步影响所接受到的光线的强度,因此,光线在云中的传播是一个复杂而反复的过程。在光线传播的简化过程中,产生了多重散射和单向散射两种理论。在本文的研究过程中,在不失真实性的前提下,将其进一步简化,采用前向为主的多重散射和单向散射相结合的方法进行光线的渲染。在光线由光源传播到云的过程中,采用前向为主的多重散射模型,即真实的表现出了光线在云中的各种传播路径,又由于对其产生的极小影响的光线进行了弱化忽略,提高了计算速度;在计算光线由云团传播到视角的过程中,采用单向散射技术,满足了真实性和速度两方面的需求。
本文中采用元胞自动机模型进行模拟,结合云的各种物理特性,从而得到了动态的,真实感强的自然云的模拟,在真实感和速度上均取得了比较满意的结果。
Over the past decade, one of the important fields in the study of Computer Graphics is the imitation of natural phenomenon, rendering of natural scene is a hot topic in the research area of graphics. How to construct complex structures, and morphological changes of the object is a new topic in computer graphics with the imitation of clouds as its hot topic. The traditional set of modeling methods can not effectively describe shapes of the clouds that have no rules, and can dynamic changes. In this paper we bases on cellular automata model and the description of clouds-like objects in hydrodynamics to further study the cloud modeling, dynamic simulation, and the illumination model. The following aspects are studied in this paper:
1. model of clouds on the basis of cellular automation.
The cellular automation is simple and high efficient. The application of cellular automation to the setting of clouds model can basically increase the speed. This paper proposed state transition rules of clouds, based on cellular automata. The rules base on the characteristics of cloud changes in the real world, and control the speed of the formation of clouds, so the model can meet the natural cloud’s state transition laws.
2. Combine with Mathematics Physics Equation to realize the natural cloud.
After studying the fluid mechanics, which describes cloud-like objects that has no rules, we extracted some natural factors that affect the movement of clouds and their affecting ways. We proposed the cloud simulation method of using of cellular automaton model to stimulate the dynamic phenomena of the cloud. This method can be combined with the natural physics of cloud movement and the using of cellular automata rules helps to improve simulation realism. In the real world, cloud height will affect the size of the wind which will affect the passive movement of the clouds; and the external environment such as the concentration, temperature and convection-diffusion, will bring about internal disturbances. These changes have been studied in mathematical physics equations. This paper uses the equations of mathematical physics to build state transition rules based on cellular automata. So the dynamic simulation of the cloud is real and speedy.
3. A deep study on the illumination rendering of natural clouds.
In nature, light propagation will reflect, refract, diffract, scatter in the medium and it is the same when it’s in the cloud. The cloud is made up of small particles, and all particles receive light and they reflect, scatter. These phenomena will have mutual influences, and further affect the strength of their received light. So the spread of light in the clouds is a complex and iterative process. There are multiple scattering theories and single scattering theories in the process of light propagation. This paper further simplifies the process of light propagation without losing the authenticity. We mainly uses the multiple forward scattering and single scattering to render the light. In the process of the light propagates to clouds, we uses the multiple scattering models, which showed various communication paths, and because of the light was weakened and ignored, this model can improve the speed; and when we calculates the light transmits from the clouds to view, we use the one-way technology, which can meet authenticity and speed.
This paper uses cellular automata model and various physical properties of clouds, to achieve a dynamic, and realistic simulation of natural cloud, which is real and speedy.
引文
[1]徐华勋.云的动态实时仿真技术研究与实现[D].湖南:国防科学技术大学研究生院.2005, 11.
[2]姚海,鲍劲松,金烨.基于元胞自动机的云层实时模拟[J].系统仿真学报,2008,20 (11):2946-2950.
[3] Bastien Chopard,Michel Droz.物理系统的元胞自动机模拟[M].北京:清华大学出版社,2003:08.
[4]孙霞,吴自勤,黄畇.分形原理及其应用[M].合肥:中国科学技术大学出版社,2006.
[5]宣慧玉,张发.复杂系统仿真及应用[M].北京:清华大学出版社,2008.
[6]赫培峰,崔建江,潘峰.计算机仿真技术[M].北京:化学工业出版社,2009.07.
[7] www.google.cn.
[8] Taxen G.Cloud Modeling for Computer Graphics [D].Royal Institute of Technology, Stockholm,Sweden,1999.
[9] Nishita T.Display of Clouds Taking into Account Multiple Anisotropic Scattering and Sky Light[C],1996:379-386.
[10] Stam J.Stable Fluids[C].1999:121-128.
[11] Matthias Unbescheiden,Andrzdj Trembilski,Cloud Simulation in Virtual Environments[A]. IEEE Visualization Proceedings[C],1998.98-104.
[12] Ebert DS.Advanced Modeling Techniques for Computer Graphics[J].ACM Computing Surveys,1996,28(1):153-156.
[13] Unbescheiden M, Andrzej T. Cloud Simulation in Virtual Environments [A].IEEE Visuali- zation Proceedings[C], 1998.98-104.
[14]唐凯,康凤举,褚彦军.Vega中云的仿真方法[J].系统仿真学报,2005,17(9):2051-2069.
[15] Nishita T, Takao S, Katsumi T, et al. Display of the earth taking into account atmospheric scattering[A].Proceedings of SIGGRAPH 1993 [C]. Anaheim, California,1993.175-182.
[16] Dobashi Y, Nishita T, YamashitaH, et al. Modeling of Cloud from Satellite Images Using Metaballs[A]. Proceedings of the 4th Pacific Conference[C], 1998.53-60.
[17] G Y Gardner. Visual Simulation Clouds[J].Computer Graphics (S1003-9767), 1985,19(3): 279- 303.
[18] Peitgen HO,Saupe D.The Science of Fractal Images[M].New York:Springer Verlag,1988.
[19]贺怀清,刘浩瀚,刘金星,杨国庆.一种改进的立体云模拟方法[J].系统仿真学报,2008,20(10) :2620-2623.
[20]高宏俊.基于流体模型和GPU的云的实时仿真[D].大连:大连理工大学,2007.
[21] Perlin K. An Image Synthesizer[J]. ACM SIGGRAPH Computer Graphics, 1985,19(3): 287-296.
[22] Dobashi Y, Kaneda K, Hamashima H, et al. A Simple, Efficient Method for Realistic Animation of Clouds[A]. SIGGRAPH2000[C],2000.19-28.
[23]张芹,谢隽毅,吴慧中等.火焰,烟,云等不规则物体的建模方法研究综述[J].中国图象图形学报,2000,5(3):186-190.
[24] P Pimentia.J, Garbocz,C. Carter Cellular Automaton Algorithm for Surface Mass Transport duo to Curvature Gradents Simulations of Sintering[J]. Computer Mater Sci.1992,1:63-77.
[25]高成锴,王斌,许凤叶,冯林.元胞自动机在图像修补中的应用研究[J],计算机工程与应用,2008,44(11):196-198.
[26]宣慧玉,张发.复杂系统仿真及应用[M],北京:清华大学出版社,2008.
[27] Wolfram S.Universality and complexity in celluar automata[J].Physica D,1984,10(1):1-35.
[28] S. wblfram, Computation theory of cellular automata[J], Communications in Mathematical Physics,96, 15-57.
[29] G. Y. Vichniac,simulating physics with cellular automata[D], physica,1984(10):96-116.
[30] Samir Boubezari,Bozena Kaminska,A Deterministic Built In Self Test Generator Based on Cellular Automata Structures[J], IEEE Transactions on Computers,1995,44(6):805-816.
[31] Peter D. Hortensius, Robert D. McLeod, and Howard C. Card, Cellular Automata-based signature analysis for built-in self-test[J], IEEE Transactions on Computers, 1990,(39):1273- 1283.
[32] Ph.Tsalides, Cellular Automata-Based Built-InSelf-Test Structures for VLSI Systems[J], Electronics Letters,1990,27(17):1350-1352.
[33]崔异.细胞自动机研究及应用[D].成都:西南交通大学,2006.
[34]C G Langton. Artificial Life[M].In D G Langton, editor. Artificial Life, SFI Studies in the Sciences of Complexity, Addison-Wesley, CA, 1989,6: 1-47.
[35]C G Langton. Preface[M],1992.
[36]TS Ray .An evolutionary approach to synthetic biology: Zen and the art of creating life [J].Artificial Life Journal, 1994, 1(1/2):179-209.
[37]于乃功,阮晓钢.基于细胞自动机的生长系统仿真模型研究[J].计算机仿真,2005,22 (2):228-231.
[38] Dobashi Y, Kaneda K, Hamashima H, et al. A Simple, Efficient Method for Realistic Animation of Clouds[A]. SIGGRAPH2000[C],2000.19-28.
[39] K Nagel,E Raschke.Self-Organizing Criticality in Clouds Formation[J].Physica A(S0378- 4371), 1992,182(4):519-531.
[40]洪炳镕,蔡则苏,唐好选.虚拟现实及其应用[M].北京:国防工业出版社,2005.
[41]吴家鸣,张国强.流体力学[M].北京:机械工业出版社,2006.
[42]盛裴轩等.大气物理学[M],北京:北京大学出版社,2005.
[43]谷超豪,李大潜,陈恕行,郑宋穆,谭永基.数学物理方程[M].高等教育出版社.
[44]李万彪等.大气概论[M],北京:北京大学出版社,2009.
[45] J F Blinn. Light Reflection Function for Simulation of Clouds and Dusty Surface[J].Computer Graphics, 1982, 16(3): 21-29.
[46] N Max. The Simulation of Natural Phenomena Panel [J]. Computer Graphics, 1983, 17(3): 137-139.
[47] J T Kajiya, H B P Von. Ray Tracing Volume Densties[J]. Computer Graphics, 1984, 18(3): 165-173.
[48] Donald Hearn,M.Pauline Baker等.计算机图形学(第二版)[M],2002.
[49]任重,周昆.复杂环境光源下的动态场景实时绘制算法综述[J].计算机辅助设计与图形学学报,2008,20(8):957-967.
[50] Nishita T.Display of Clouds Taking into Account Multiple Anisotropic Scattering andSky Light[C]//SIGGRAPH96 Proceedings. New York:ACM,1996:379-386.
[51] Mark J.Harris, Anselmo Lastra. Real-Time Cloud Rendering[J],Computer Science,2001,20(3).
[52]徐利明,姜昱明.基于粒子系统与OpenGL的实时雨雪模拟[J],计算机仿真,2005,22(7):242-245.
[2]姚海,鲍劲松,金烨.基于元胞自动机的云层实时模拟[J].系统仿真学报,2008,20 (11):2946-2950.
[3] Bastien Chopard,Michel Droz.物理系统的元胞自动机模拟[M].北京:清华大学出版社,2003:08.
[4]孙霞,吴自勤,黄畇.分形原理及其应用[M].合肥:中国科学技术大学出版社,2006.
[5]宣慧玉,张发.复杂系统仿真及应用[M].北京:清华大学出版社,2008.
[6]赫培峰,崔建江,潘峰.计算机仿真技术[M].北京:化学工业出版社,2009.07.
[7] www.google.cn.
[8] Taxen G.Cloud Modeling for Computer Graphics [D].Royal Institute of Technology, Stockholm,Sweden,1999.
[9] Nishita T.Display of Clouds Taking into Account Multiple Anisotropic Scattering and Sky Light[C],1996:379-386.
[10] Stam J.Stable Fluids[C].1999:121-128.
[11] Matthias Unbescheiden,Andrzdj Trembilski,Cloud Simulation in Virtual Environments[A]. IEEE Visualization Proceedings[C],1998.98-104.
[12] Ebert DS.Advanced Modeling Techniques for Computer Graphics[J].ACM Computing Surveys,1996,28(1):153-156.
[13] Unbescheiden M, Andrzej T. Cloud Simulation in Virtual Environments [A].IEEE Visuali- zation Proceedings[C], 1998.98-104.
[14]唐凯,康凤举,褚彦军.Vega中云的仿真方法[J].系统仿真学报,2005,17(9):2051-2069.
[15] Nishita T, Takao S, Katsumi T, et al. Display of the earth taking into account atmospheric scattering[A].Proceedings of SIGGRAPH 1993 [C]. Anaheim, California,1993.175-182.
[16] Dobashi Y, Nishita T, YamashitaH, et al. Modeling of Cloud from Satellite Images Using Metaballs[A]. Proceedings of the 4th Pacific Conference[C], 1998.53-60.
[17] G Y Gardner. Visual Simulation Clouds[J].Computer Graphics (S1003-9767), 1985,19(3): 279- 303.
[18] Peitgen HO,Saupe D.The Science of Fractal Images[M].New York:Springer Verlag,1988.
[19]贺怀清,刘浩瀚,刘金星,杨国庆.一种改进的立体云模拟方法[J].系统仿真学报,2008,20(10) :2620-2623.
[20]高宏俊.基于流体模型和GPU的云的实时仿真[D].大连:大连理工大学,2007.
[21] Perlin K. An Image Synthesizer[J]. ACM SIGGRAPH Computer Graphics, 1985,19(3): 287-296.
[22] Dobashi Y, Kaneda K, Hamashima H, et al. A Simple, Efficient Method for Realistic Animation of Clouds[A]. SIGGRAPH2000[C],2000.19-28.
[23]张芹,谢隽毅,吴慧中等.火焰,烟,云等不规则物体的建模方法研究综述[J].中国图象图形学报,2000,5(3):186-190.
[24] P Pimentia.J, Garbocz,C. Carter Cellular Automaton Algorithm for Surface Mass Transport duo to Curvature Gradents Simulations of Sintering[J]. Computer Mater Sci.1992,1:63-77.
[25]高成锴,王斌,许凤叶,冯林.元胞自动机在图像修补中的应用研究[J],计算机工程与应用,2008,44(11):196-198.
[26]宣慧玉,张发.复杂系统仿真及应用[M],北京:清华大学出版社,2008.
[27] Wolfram S.Universality and complexity in celluar automata[J].Physica D,1984,10(1):1-35.
[28] S. wblfram, Computation theory of cellular automata[J], Communications in Mathematical Physics,96, 15-57.
[29] G. Y. Vichniac,simulating physics with cellular automata[D], physica,1984(10):96-116.
[30] Samir Boubezari,Bozena Kaminska,A Deterministic Built In Self Test Generator Based on Cellular Automata Structures[J], IEEE Transactions on Computers,1995,44(6):805-816.
[31] Peter D. Hortensius, Robert D. McLeod, and Howard C. Card, Cellular Automata-based signature analysis for built-in self-test[J], IEEE Transactions on Computers, 1990,(39):1273- 1283.
[32] Ph.Tsalides, Cellular Automata-Based Built-InSelf-Test Structures for VLSI Systems[J], Electronics Letters,1990,27(17):1350-1352.
[33]崔异.细胞自动机研究及应用[D].成都:西南交通大学,2006.
[34]C G Langton. Artificial Life[M].In D G Langton, editor. Artificial Life, SFI Studies in the Sciences of Complexity, Addison-Wesley, CA, 1989,6: 1-47.
[35]C G Langton. Preface[M],1992.
[36]TS Ray .An evolutionary approach to synthetic biology: Zen and the art of creating life [J].Artificial Life Journal, 1994, 1(1/2):179-209.
[37]于乃功,阮晓钢.基于细胞自动机的生长系统仿真模型研究[J].计算机仿真,2005,22 (2):228-231.
[38] Dobashi Y, Kaneda K, Hamashima H, et al. A Simple, Efficient Method for Realistic Animation of Clouds[A]. SIGGRAPH2000[C],2000.19-28.
[39] K Nagel,E Raschke.Self-Organizing Criticality in Clouds Formation[J].Physica A(S0378- 4371), 1992,182(4):519-531.
[40]洪炳镕,蔡则苏,唐好选.虚拟现实及其应用[M].北京:国防工业出版社,2005.
[41]吴家鸣,张国强.流体力学[M].北京:机械工业出版社,2006.
[42]盛裴轩等.大气物理学[M],北京:北京大学出版社,2005.
[43]谷超豪,李大潜,陈恕行,郑宋穆,谭永基.数学物理方程[M].高等教育出版社.
[44]李万彪等.大气概论[M],北京:北京大学出版社,2009.
[45] J F Blinn. Light Reflection Function for Simulation of Clouds and Dusty Surface[J].Computer Graphics, 1982, 16(3): 21-29.
[46] N Max. The Simulation of Natural Phenomena Panel [J]. Computer Graphics, 1983, 17(3): 137-139.
[47] J T Kajiya, H B P Von. Ray Tracing Volume Densties[J]. Computer Graphics, 1984, 18(3): 165-173.
[48] Donald Hearn,M.Pauline Baker等.计算机图形学(第二版)[M],2002.
[49]任重,周昆.复杂环境光源下的动态场景实时绘制算法综述[J].计算机辅助设计与图形学学报,2008,20(8):957-967.
[50] Nishita T.Display of Clouds Taking into Account Multiple Anisotropic Scattering andSky Light[C]//SIGGRAPH96 Proceedings. New York:ACM,1996:379-386.
[51] Mark J.Harris, Anselmo Lastra. Real-Time Cloud Rendering[J],Computer Science,2001,20(3).
[52]徐利明,姜昱明.基于粒子系统与OpenGL的实时雨雪模拟[J],计算机仿真,2005,22(7):242-245.