Design and Implementation of Cluster-Based Platform for Remote Sensing Information Computation
Remote Sensing Information Computation has features of large amount of image data and complex algorithms, many applications require real-time response, but mainstream computers are often unable to meet their performance requirements. Regarding to this problem, a cluster-based computing platform for remote sensing information computation was designed and implemented through interconnecting the existing computing devices by network, and in support of MPI and other communication software, cluster management module which enables node management, task management, and realtime monitoring and other functions was developed. On this basis, the parallel algorithms of re-projection, mean-shift multi-scale segmentation were achieved; the efficiency and performance of the cluster were tested. Experiment results show that the system can significantly improve the efficiency of remote sensing information computation with a low price.
remote sensing computation cluster MPI
Wang Weihong Chen Jiechen Gu Guomin Wu Wei
College of Computer Science, Zhejiang University of Technology Hangzhou, 310023, China Remote Sensing Applications, Chinese Academy of Sciences Beijing, 100101, China
国际会议
哈尔滨
英文
2122-2125
2011-12-24（万方平台首次上网日期，不代表论文的发表时间）