全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于MPI的大规模遥感影像金字塔并行构建方法

DOI: 10.3724/SP.J.1047.2015.00515, PP. 515-522

Keywords: 集群,遥感影像金字塔,消息传递接口,并行

Full-Text   Cite this paper   Add to My Lib

Abstract:

影像金字塔是实现影像数据多分辨率组织的重要方式,是提高影像可视化性能的有效手段。传统串行金字塔构建算法,对大规模影像数据的构建性能已无法满足遥感影像快速浏览的预处理需求。故此,其成为一个亟待解决的问题,而利用多核、多节点的高性能集群计算环境和并行机制是一个重要的技术途径。本文在共享外存的高性能集群环境下,提出使用消息传递接口(MPI)的金字塔并行构建算法,对构建遥感影像金字塔过程中的重采样与I/O过程进行并行处理,大大缩短了遥感影像金字塔构建时间。实验结果表明(1)该算法比传统串行构建方法的加速效果明显,对于单波段遥感影像,其加速效果可达到GDAL的5倍以上,而对于多波段遥感影像,加速效果可达到GDAL的2倍以上;(2)遥感影像数据量越大,并行构建算法加速效果越显著,对于大规模的遥感影像,本文提出的金字塔并行构建算法的速度可达到GDAL的10倍左右。

References

[1]  苏光日,范义鹏,张学之,等.GDAL 在地理国情监测项目 中的应用[J].测绘与空间地理信息,2014,37(1):86-88.
[2]  Qin C Z, Zhan L J, Zhu A. How to apply the Geospatial Data Abstraction Library (GDAL) properly to parallel geospatial raster I/O?[J]. Transactions in GIS, 2014,18(6): 950-957.
[3]  Zhang S Q, Chen L, Xiong W. Research on performances of parallel programming models based on chip multi-processor[C]. 2012 IEEE International Conference on Oxide Materials for Electronic Engineering (OMEE), 2012:632-635.
[4]  Wang S. CyberGIS: Blueprint for integrated and scalable geospatial software ecosystems[J]. International Journal of Geographical Information Science, 2013,27(11):2119-2121.
[5]  Yang C, Goodchild M, Huang Q, et al. Spatial cloud computing: how can the geospatial sciences use and help shape cloud computing?[J]. International Journal of Digital Earth, 2011,4(4):305-329.
[6]  康俊锋,杜震洪,刘仁义,等.基于GPU加速的遥感影像金 字塔创建算法及其在土地遥感影像管理中的应用[J].浙 江大学学报:理学版,2011,38(6):695-700.
[7]  Chen G D, Yang J Y. Research on orthorectification of remote sensing images using GPU-CPU cooperative processing[C]. 2011 International Symposium on Image and Data Fusion (ISIDF), IEEE, 2011:1-4.
[8]  刘小利,徐攀登,朱国宾,等.结合MapReduce 和HBase 的 遥感图像并行分布式查询[J].地理与地理信息科学, 2014,30(5):26-28.
[9]  刘义,陈荦,景宁,等.利用MapReduce 进行批量遥感影像 瓦片金字塔构建[J].武汉大学学报:信息科学版,2013(3): 278-282.
[10]  Ma Y, Liu D, Li J. A new framework of cluster-based parallel processing system for high-performance geo-computing[C]. IEEE International Geoscience and Remote Sensing Symposium, 2009,4:IV49-IV52.
[11]  Guo W, Zhu X, Hu T, et al. A Multi-granularity parallel model for unified remote sensing image processing Web-Services[J]. Transactions in GIS, 2012,16(6):845-866.
[12]  Guan X, Wu H, Li L. A parallel framework for processing massive spatial data with a split-and-merge paradigm[J]. Transactions in GIS, 2012,16(6):829-843.
[13]  Thakur R, Gropp W, Lusk E. Optimizing noncontiguous accesses in MPI-IO[J]. Parallel Computing, 2002,28(1): 83-105.
[14]  Didelot S, Carribault P, Pérache M, et al. Improving MPI communication overlap with collaborative polling[J]. Computing, 2014,96(4):263-278.
[15]  申焕,石晓春,邱宏华.基于MPI 的海量遥感影像并行处 理技术探析[J].全球定位系统,2013,37(6):73-76.
[16]  欧阳柳,熊伟,程果,等.地理栅格数据的并行访问方法研 究[J].计算机科学,2012,39(11):116-121.
[17]  Porwal S, Katiyar S K. Performance evaluation of various resampling techniques on IRS imagery[C]. 2014 Seventh International Conference on Contemporary Computing (IC3), IEEE, 2014:489-494.
[18]  Parker J A, Kenyon R V, Troxel D. Comparison of interpolating methods for image resampling[J]. IEEE Transactions on Medical Imaging, 1983,2(1):31-39.
[19]  Chen Y, Sun X H, Thakur R, et al. Improving parallel I/O performance with data layout awareness[C]. 2010 IEEE International Conference on Cluster Computing (CLUSTER), 2010:302-311.
[20]  Ferhatosmanoglu H, Agrawal D, Egecioglu, et al. Optimal data-space partitioning of spatial data for parallel I/O[J]. Distributed and Parallel Databases, 2005,17(1):75-101.
[21]  Liu O Y, Huang J L, Wu X H, et al. Parallel access optimization technique for geographic raster data[A]. In: Bian F L, Xie Y C, Cui X H, et al (eds.). Geo-Informatics in Resource Management and Sustainable Ecosystem[C]. Springer Berlin Heidelberg, 2013:533-542.
[22]  杨伟光,李文.使用MPI的并行I/O 实现及性能分析[J].计 算机工程与应用,2006,42(17):96-98. 521

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133