长江科学院院报 ›› 2014, Vol. 31 ›› Issue (2): 91-96.DOI: 10.3969/j.issn.1001-5485.2014.02.019

• 信息技术应用 • 上一篇    下一篇

基于MapReduce的高分辨率遥感影像特征提取方法

沈盛彧1, 刘哲2, 张平仓1, 张彤3, 吴华意3, 陈小平1   

  1. (1.长江科学院 水土保持研究所,武汉 430010;2. 长江水利委员会 网络与信息中心,武汉 430010;3.武汉大学 测绘遥感信息工程国家重点实验室,武汉 430079)
  • 收稿日期:2013-02-04 修回日期:2013-07-05 出版日期:2014-01-26 发布日期:2014-01-27
  • 作者简介:沈盛彧(1984- ),男,湖北武汉人,工程师,博士,主要从事高分辨遥感影像处理与水土保持研究,(电话)027-82926365(电子信箱)shshy.whu@gmail.com。
  • 基金资助:
    国家自然科学基金资助项目(41271400);国家973计划资助项目(2012CB719906);中央级公益性科研院所基本科研业务费(CKSF2012044TB,CKSF2012055TB)

Extraction of High-resolution Remote-sensing Image Feature Based on MapReduce

SHEN Sheng-yu1,LIU Zhe2,ZHANG Ping-cang1,ZHANG Tong3,WU Hua-yi3,CHEN Xiao-ping1   

  1. (1.Soil and Water Conservation Department, Yangtze River Scientific Research Institute,Wuhan 430010, China; 2. Network and Information Center, Changjiang Water Resources Commission,Wuhan 430010, China; 3. State Key Laboratory of Information Engineering in Surveying, Mapping,and Remote Sensing, Wuhan University, Wuhan 430079, China)
  • Received:2013-02-04 Revised:2013-07-05 Online:2014-01-26 Published:2014-01-27

摘要: 遥感影像的数量和数据量正在呈几何级数增长,传统遥感影像处理方法已经无法应对这一海量问题。利用新兴的高性能计算集群的超强计算、存储及吞吐能力处理海量高分辨率遥感影像是一种新的思路。在基于云计算的高分辨率遥感影像处理的研究框架下,介绍一种MapReduce遥感影像特征提取方法,实现海量高分辨率遥感影像的海量底层视觉特征的提取。通过在16个节点的Hadoop集群上进行数据量扩展和处理能力扩展实验,证明了基于MapReduce的高分辨率遥感影像底层视觉特征的高效检测与描述方法的高效率及可扩展性。

关键词: 云计算, 高分辨率遥感影像, 底层视觉特征, MapReduce

Abstract: Since the number and amount of remote sensing images is growing exponentially, traditional sensing image processing methods have been unable to deal with this massive growth. The supercomputing, massive storage and handling capacity of the high-performance computing cluster is a new solution to deal with the massive high-resolution remote sensing images. A method of extracting the basic visual features of high-resolution remote sensing images based on MapReduce is proposed. By experiments on the expansion of data amount and processing capacity on a 16-node Hadoop cluster, the MapReduce-based method is proved to be effective and scalable.

Key words: cloud computing, high-resolution remote sensing image, basic visual features, MapReduce

中图分类号: