应用科学学报 ›› 2017, Vol. 35 ›› Issue (5): 658-666.doi: 10.3969/j.issn.0255-8297.2017.05.011

• 信号与信息处理 • 上一篇    下一篇

一种多光谱遥感应用需求综合方法

巫兆聪, 刘培, 巫远   

  1. 武汉大学 遥感信息工程学院, 武汉 430079
  • 收稿日期:2016-08-19 修回日期:2016-11-30 出版日期:2017-09-30 发布日期:2017-09-30
  • 作者简介:巫兆聪,教授,博导,研究方向:高分辨率遥感影像信息提取、遥感数据智能处理、定量遥感,卫星仿真与效能评估,E-mail:zcwoo@163.com
  • 基金资助:

    民用航天“十二·五”预先研究项目基金(No.20136697)资助

Synthesis of Requirements in Applications of Multi-spectral Remote Sensing

WU Zhao-cong, LIU Pei, WU Yuan   

  1. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China
  • Received:2016-08-19 Revised:2016-11-30 Online:2017-09-30 Published:2017-09-30

摘要:

中国各遥感应用需求部门对于需求的描述存在很大差别,而且需求之间有一定相关性,给后续遥感传感器的设计带来很大不便.针对这个问题,以多光谱应用需求为例,提出一种需求综合方法.首先将相关部门的原始需求以基于领域本体建模的方法进行结构化表达,建立相应的需求数据库以保证数据共享性,然后构建多光谱应用需求空间分辨率–时间分辨率的需求分类决策树.与中国空间基础设施规划的对地观测多光谱应用需求进行比较,并对分类方法进行评估.结果表明,该方法能获取良好的综合后需求,可直接用于后续遥感传感器的设计与评估.

关键词: 遥感对地观测, 决策树分类, 多光谱应用需求, 领域本体建模

Abstract:

Great differences in requirement descriptions exist in remote sensing applications, and various requirements may be correlated. This causes inconvenience for the subsequent sensor design, We propose a method to solve the problem for multi-spectral remote sensing application requirements. We frst express the original requirements of the associated departments in a structural manner based on the domain ontology modeling, and establish a database to ensure data sharing. We then establish a spatial resolution- temporal resolution decision tree to classify and eliminate correlation, and evaluate the results. We show that good synthesized requirements can be obtained, and used for subsequent sensor design for earth-oriented observation remote sensing, and it can be evaluated directly.

Key words: multi-spectral application requirement, decision tree classifcation, remote sensing earth-oriented observation, domain ontology modeling

中图分类号: