异构环境下顺序无关的图像合成算法

系统仿真学报 ›› 2015, Vol. 27 ›› Issue (9): 2117-2125.

异构环境下顺序无关的图像合成算法

刘宁^1,2, 朱登明¹, 王兆其¹, 石敏^1,3

1.中国科学院计算技术研究所,北京 100190;
2.中国科学院大学,北京 100049;
3.华北电力大学,北京 102206

收稿日期:2015-06-04 修回日期:2015-07-02 出版日期:2015-09-08 发布日期:2020-08-07
作者简介:刘宁(1987-),男,山东,博士生,研究方向为科学数据可视化;朱登明(1973-),男,博士,副研究员,研究方向为自然现象模拟和科学数据可视化;王兆其(1966-),男,博士,研究员,研究方向为虚拟现实和智能人机交互。
基金资助:
国家自然科学基金(61173067; 61379085; 61300131)

Order-independent Image Compositing in Heterogeneous Environments

Liu Ning^1,2, Zhu Dengming¹, Wang Zhaoqi¹, Shi Min^1,3

1. Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China;
2. University of Chinese Academy of Sciences, Beijing 100049, China;
3. North China Electric Power University, Beijing 102206, China

Received:2015-06-04 Revised:2015-07-02 Online:2015-09-08 Published:2020-08-07

摘要/Abstract

摘要： 传统的图像合成算法是为同构环境设计的。对于异构环境下顺序无关的图像合成问题,这些算法不考虑环境的异构性,也没有利用顺序无关合成操作符的性质,容易造成资源的浪费。针对这个问题,提出了一个异构环境下顺序无关的图像合成算法,它会根据环境的计算资源与网络资源更合理地进行任务分配,并利用顺序无关合成操作符的性质来优化合成方案。在ns-3网络模拟器上进行了大量的实验。实验结果表明:当环境存在较大异构性时,该方法相对经典算法binary swap能够取得显著的加速比。

关键词: 图像合成, 异构环境, 并行可视化, 差分进化算法, ns-3网络模拟器

Abstract: Conventional image compositing algorithms are designed for homogeneous environments. These algorithms do not take heterogeneity into account and do not make use of the properties of an order-independent compositing operator. As for order-independent image compositing problems in a heterogeneous environment, these algorithms usually can not make full use of the resources. An order-independent image compositing algorithm in heterogeneous environments was proposed, which was able to distribute the workload according to computing and networking characteristics, and managed to optimize compositing scheme by utilizing the properties of order-independent operator. A comprehensive evaluation on the ns-3 network simulator was conducted. Experiment results show when the environment has great heterogeneity, the algorithm will gain significant speedup over classical algorithm binary swap.

Key words: image compositing, heterogeneous environment, parallel visualization, differential evolution, ns-3 simulator

中图分类号:

TP391

刘宁, 朱登明, 王兆其, 石敏. 异构环境下顺序无关的图像合成算法[J]. 系统仿真学报, 2015, 27(9): 2117-2125.

Liu Ning, Zhu Dengming, Wang Zhaoqi, Shi Min. Order-independent Image Compositing in Heterogeneous Environments[J]. Journal of System Simulation, 2015, 27(9): 2117-2125.

参考文献

[1] Neumann U.Parallel volume-rendering algorithm performance on mesh-connected multicomputers [C]// Proceedings of the Symposium on Parallel Rendering. New York, USA: ACM Press, 1993: 97-104.
[2] Ma K L, Painter J, Hansen C, et al. Parallel volume rendering using binary-swap compositing[J]. IEEE Computer Graphics and Applications(S0272-1716) , 1994, 14(4): 59-68.
[3] Yu H, Wang C, Ma K L.Massively parallel volume rendering using 2-3 swap image compositing [C]// Proceedings of the ACM/IEEE Conference on Supercomputing. New Jersey, USA: IEEE Press, 2008, 48:1-48:11
[4] Peterka T, Goodell D, Ross R, et al.A configurable algorithm for parallel image-compositing applications [C]// Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis. New York, USA: ACM Press, 2009: 4: 1-4:10
[5] Wu Q, Gao J, Chen Z, et al. Pipelining parallel image compositing and delivery for efficient remote visualization[J]. Journal of Parallel and Distributed Computing(S0743-7315), 2009, 69(3): 230-238.
[6] R Storn, K Price.Differential evolution a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of Global Optimization(S0925-5001), 1997, 11(4): 341-359.
[7] J C Bean.Genetic algorithms and random keys for sequencing and optimization[J]. ORSA Journal On Computing(S0899-1499), 1994, 6(2): 154-160.
[8] Ns-3 developers. ns-3 network simulator [EB/OL]. (2015-05-01). http://www.nsnam.org/
[9] Kendall W, Peterka T, Huang J, et al. Accelerating and benchmarking radix-k image compositing at large scale[C]// Proceedings of the Eurographics Conference on Parallel Graphics and Visualization. Aire-la-Ville, France: Eurographics Association, 2010: 101-110.
[10] Moreland K, Kendall W, Peterka T, et al.An image compositing solution at scale [C]// Proceedings of International Conference for High Performance Computing, Networking, Storage and Analysis. New York, USA: ACM Press, 2011, 25: 1-25: 10.

[1]	余伟, 梁恒辉, 罗映. 一种改进差分进化算法的分数阶系统辨识研究[J]. 系统仿真学报, 2021, 33(5): 1157-1166.
[2]	梅盼盼, 吴亮红, 张红强, 王慧莹. 含风电电力系统动态经济调度的差分纵横交叉优化[J]. 系统仿真学报, 2020, 32(6): 1179-1187.
[3]	王艳, 丁宇. 动态柔性作业车间优化调度与决策方法[J]. 系统仿真学报, 2020, 32(11): 2073-2083.
[4]	林雨谷, 王艳. 离散车间能效数据挖掘及调度优化[J]. 系统仿真学报, 2019, 31(12): 2702-2711.
[5]	王杰, 陈彬, 袁鹏, 马亮, 邱晓刚. 数据驱动的最优互惠避碰模型偏好速度研究[J]. 系统仿真学报, 2019, 31(12): 2731-2739.
[6]	张静华, 韩璞. 多算法多种群协同优化差分进化算法[J]. 系统仿真学报, 2018, 30(5): 1690-1699.
[7]	陈文杰, 王艳. 离散制造过程生产协同能耗优化调度方法[J]. 系统仿真学报, 2018, 30(11): 4367-4375.
[8]	李芹, 张浩, 彭道刚, 郭义波, 王念龙, 孙宇贞. 基于改进差分算法的电站锅炉主蒸汽温度多变量建模研究[J]. 系统仿真学报, 2017, 29(8): 1712-1718.
[9]	张妍, 王东风, 韩璞. 基于IDE-LSSVM的风电场短期风速预测[J]. 系统仿真学报, 2017, 29(7): 1561-1571.
[10]	刘增良, 周松林, 周同旭. 基于IDE-WNN的短期风电功率预测及概率评估[J]. 系统仿真学报, 2016, 28(2): 476-482.
[11]	徐小平, 白博, 钱富才. 基于改进差分进化算法的Wiener模型辨识[J]. 系统仿真学报, 2016, 28(1): 147-153.
[12]	彭琛, 钱晓山. 基于聚类的多模型蒸发过程软测量建模[J]. 系统仿真学报, 2015, 27(9): 2050-2055.
[13]	孙荣玲, 钱晓山. 基于约简鲁棒LSSVM铝酸钠浓度软测量[J]. 系统仿真学报, 2015, 27(9): 2203-2207.
[14]	刘宁, 朱登明, 王兆其, 魏毅. 异构环境下基于分组的图像合成算法[J]. 系统仿真学报, 2015, 27(8): 1801-1808.