基于Hadoop的地表温度反演系统的设计与实现  

Design and implementation of land surface temperature retrieving system based on Hadoop

作　　者：袁帅 郑逢斌[1] Yuan Shuai, Zheng Fengbin(School of Computer and Information Engineering Henan University, Kaifeng, Henan 475004, Chin)

机构地区：[1]河南大学计算机与信息工程学院,河南开封475004

出　　处：《计算机时代》2018年第5期32-35,40,共5页Computer Era

基　　金：高分辨率卫星遥感影像灰霾颗粒物反演算法与估算模型研究(41571417)

摘　　要：随着我国空间技术的发展和遥感信息模型复杂化,卫星遥感的数据量呈指数形快速增长。在地表温度反演过程中,进行海量遥感数据的处理和信息提取成了不可回避的现实。本系统以云计算框架,Hadoop平台为基础,应用HDFS和Map Reduce等技术实现了对海量遥感数据的冗余备份、信息存储和并行计算,以便于更高效地获得地表温度值。通过对高分卫星数据的处理和反演产品生产过程的分析,证明本系统具有更高的运行计算能力和获取结果效率。With the development of space technology and the complication of remote sensing information model in China, the data of satellite remote sensing has increased rapidly in an exponential form. In the process of surface temperature inversion, the processing and information extraction of massive remote sensing data has become an unavoidable reality. Based on cloud computing framework and Hadoop platform, HDFS and Map Reduce technologies are applied to realize redundant backup,information storage and parallel computing for a large number of remote sensing data, so as to obtain land surface temperature more efficiently. By analyzing the processing of high resolution satellite data and the process of obtaining the temperature inversion product, it is found that the system has higher computing ability and efficiency of obtaining results.

关 键 词：HADOOP 海量信息存储 并行计算 地表温度反演 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]