首页 | 本学科首页   官方微博 | 高级检索  
     检索      

云环境中基于金字塔模型的影像数据存储方法
引用本文:夏英,杨选伦.云环境中基于金字塔模型的影像数据存储方法[J].重庆邮电大学学报(自然科学版),2012,24(6):669-674.
作者姓名:夏英  杨选伦
作者单位:重庆邮电大学计算机科学与技术学院,重庆,400065
基金项目:国家自然科学基金(41101432);重庆市自然科学基金(CSTC,2010BB2416)
摘    要:随着卫星遥感获取技术的发展,影像数据量呈几何式增长,传统数据存储技术已经无法适应海量影像数据的处理要求.结合金字塔模型和MapReduce架构提出了一种适用于云计算环境的分布式并行存储方法—BMLStorage(storage based on MapReduce and local file system),该方法基于金字塔模型对影像数据进行分层分块处理,并对所得瓦片重新编码.定义了一种新的存储规则,使得所有瓦片可以按照此规则利用Hadoop中的MapReduce框架实现并行存储.实验表明,该方法与现有方法相比,在海量影像数据存储性能方面有一定的提高.

关 键 词:云计算  MapReduce  金字塔模型  影像数据
收稿时间:2012/2/27 0:00:00

Image data storage method based on pyramid model in cloud
XIA Ying,YANG Xuanlun.Image data storage method based on pyramid model in cloud[J].Journal of Chongqing University of Posts and Telecommunications,2012,24(6):669-674.
Authors:XIA Ying  YANG Xuanlun
Institution:School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P.R.China
Abstract:With the development of satellite remote sensing technology, the size of image data is growing at an incredible rate, so the traditional data storage technology can not satisfy the massive image data processing requirements. In this paper, the BML Storage (storage based on MapReduce and local file system) method is proposed for improving the performance of massive image data storage. It is a distributed parallel storage method combined with Pyramid model, MapReduce thinking and cloud computing. This method recodes image data which has been stratified and blocked based on the pyramid model, and a storage rule is defined to ensure that all of the tiles can be easily stored inparallel. Compared with existing methods, experiments show that the performance of massive image data storage is improved in a certain extent.
Keywords:cloud computing  MapReduce  pyramid model  image data
本文献已被 万方数据 等数据库收录!
点击此处可从《重庆邮电大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《重庆邮电大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号