基于云平台的遥感业务流程研究及原型实现

发布时间：2018-11-26 13:01

【摘要】：随着航空航天技术和卫星传感器的发展，遥感技术受到国家越来越多的重视，对地观测遥感卫星的应用规模也在不断地扩大。并且遥感影像分辨率的不断提高导致了遥感影像的数据量成几何倍数激增。因此，如何对这些不断增长的海量遥感影像数据进行计算、管理以及存储，并在此基础上提供高效的计算服务已经成为了一个当今迫切需要解决的难题。而随着计算机技术的发展，云计算的诞生为大数据量存储以及计算提供了解决方案。云计算在目前并没有给出一个官方的概念，但普遍被认同是网络技术和多种计算机技术，包括分布式计算、并行计算、效用计算混合发展的结果。云计算拥有可扩展性、超大规模、可靠性、高可用性以及经济廉价等特点，能够聚集闲置的异构资源，实现大规模计算的能力，可以作为一个高效解决大数据量遥感影像处理与管理的有效手段。本文在国内外对云计算、遥感影像计算服务和共享存储的研究基础上，总结了云计算、工作流以及SOA等关键理论与技术，探索了如何利用云计算平台的计算能力与存储能力实现对大量遥感影像的共享存储与快速批量的并行处理。论文重点对BPEL与Platform PM两种工作流引擎的工作机制进行了研究，并在此基础上设计和实现了遥感业务流程原型系统，能向外提供高效的遥感影像模型的计算服务、存储服务以及流程管理服务。最后，以常用的遥感指数模型NDVI到VCI计算的简单流程作为实例进行了平台测试。研究与测试结果表明，本文提出的结合云计算、SOA架构和工作流技术，，利用BPEL与Platform PM两种工作流引擎，实现的工作流原型系统可以对大数量遥感影像数据进行高效地处理和管理，体现了云计算对遥感影像的高性能计算能力和协同处理能力，为云计算在大数据量的遥感模型计算中的应用提供了参考模式。
[Abstract]:With the development of aerospace technology and satellite sensors, more and more attention has been paid to remote sensing technology, and the application scale of Earth observation remote sensing satellite is also expanding. And the increasing resolution of remote sensing image leads to the geometric multiple of remote sensing image data. Therefore, how to compute, manage and store these massive remote sensing image data and provide efficient computing services has become an urgent problem. With the development of computer technology, cloud computing provides a solution for mass data storage and computing. Cloud computing has not given an official concept at present, but it is generally recognized as the result of the mixed development of network technology and various computer technologies, including distributed computing, parallel computing, and utility computing. Cloud computing has the characteristics of scalability, large scale, reliability, high availability and low cost. It can gather idle heterogeneous resources and realize the ability of large-scale computing. It can be used as an efficient method to solve the problem of large amount of remote sensing image processing and management. Based on the research of cloud computing, remote sensing image computing service and shared storage at home and abroad, this paper summarizes the key theories and technologies of cloud computing, workflow and SOA. This paper explores how to use the computing power and storage capacity of cloud computing platform to realize the shared storage and fast batch parallel processing of a large number of remote sensing images. This paper focuses on the working mechanism of BPEL and Platform PM workflow engine, and designs and implements a remote sensing business process prototype system, which can provide efficient remote sensing image model computing services. Storage services and process management services. Finally, the simple process of NDVI to VCI calculation is used as an example to test the platform. The research and test results show that this paper combines cloud computing, SOA architecture and workflow technology, and uses two kinds of workflow engine, BPEL and Platform PM. The workflow prototype system can efficiently process and manage the large amount of remote sensing image data, which reflects the high performance computing ability and collaborative processing ability of cloud computing to remote sensing image. It provides a reference model for the application of cloud computing in the remote sensing model calculation of large amount of data.
【学位授予单位】：中国地质大学（北京）
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP3;TP79

【参考文献】