基于COG、Hadoop和Spark的海量影像快速可视化共享方法  

Fast visual sharing method of massive images based on COG,Hadoop and Spark

作　　者：杨立业 YANG Li-Ye(Fundamental Geographic Information Center of Gansu Province, Lanzhou Gansu 730000, China)

机构地区：[1]甘肃省基础地理信息中心,甘肃兰州730000

出　　处：《河北省科学院学报》2022年第3期25-31,共7页Journal of The Hebei Academy of Sciences

摘　　要：随着遥感技术的飞速发展,海量遥感影像及时、高效、波段可灵活组合可视化共享需求强烈,针对影响海量影像快速可视化共享服务显示速度及预处理时间长的关键问题,提出了基于COG(Cloud Optimized GeoTIFF)技术、Hadoop分布式存储技术和Spark并行运算技术的解决方案,解决了传统金字塔栅格瓦片时效性差、不支持波段组合显示等问题,已在项目和实际生产服务中得到应用验证。With the rapid development of remote sensing technology,there is a strong demand for timely,efficient and flexible combination of bands of massive remote sensing images.Aiming at key problems affecting the display speed and long preprocessing time of massive image fast visualization sharing service,the author proposes the solution which are based on Cloud Optimized GeoTIFF technology,Hadoop distributed storage technology and spark parallel computing technology.It can solve problems that the traditional pyramid grid tile has poor timeliness and does not support band combination display,which has been applied and verified in projects and actual production services.

关 键 词：Cloud Optimized GeoTIFF HDFS分布式存储 SPARK Scala语言 可视化显示 波段组合 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]