节点文献

基于SPOT5土地利用信息提取技术研究

The Picking-up Land Use Data by SPOT5

【作者】 马文杰

【导师】 章孝灿;

【作者基本信息】 浙江大学 , 地球探测与信息技术, 2008, 硕士

【摘要】 本文是结合浙江省科技计划项目——基于SPOT5数据水土流失遥感动态监测技术研究(2006C33056)而展开的,由于土地利用信息是影响水土流失的最重要因子之一,本文主要研究基于SPOT5影像数据获取土地利用信息的技术。为了较好的获取土地利用信息,本文结合SPOT5影像几何分辨率特点采用了一种基于知识的分层分类方法。在对遥感影像进行特征提取时,结合不同地物自身的波谱特性应用了不同的特征提取方法,进行水体特征提取时,使用了波段运算的方法得到水体特征波段PRWI,针对茶树使用波形判断与单波段阈值控制相结合的方法确定分类条件,并设计分类器。通过2005年11月余杭区SPOT5影像的具体分类实验,结合野外实地勘测的土地利用数据,进行分类结果精度评价,证明得到的土地利用类型精度是较高的,能够满足水土流失提取模型的土地利用类型精度要求,表明分层分类方法是比较适合水土流失提取模型中的土地利用类型获取的。并对比了监督分类的结果,大大提高了遥感影像分类的精度。同时本文尽可能的把分类细化到具体的树种(茶树),本次试验方法为相似的研究提供了可行的思路。通过本文的研究,为下一步提取水土流失信息奠定了良好的基础。

【Abstract】 This article is one part of scientific plan project of ZhejiangProvince----The technology research of dynamic monitoring water and soilerosion based on SPOT5 data by remote sensing(2006C33056). Study the technology of extracting the land use type by SPOT5 image data which is one of most important factors to influencing water and soil erosion.In order to pick up high accuracy land use data , the article combines the high space resolution characteristic of SPOT5 image data to design a new method which makes Classification Method of Remote Sensing Image on Basis of Knowledge and Sub-Region Classification Method together , and makes it use in this study .At the point of picking up character on the remote sensing image , in order to obtain character wave band of the water-PRWI , this author use wave band operation , and reckon classification terms to design the classification machine of the tea , which uses a method that combines comparing wave band shape and limiting threshold value rang .By concrete classification experiment of Yu Hang District by SPOT5 image in November, 2005 and combining land use data which measured on the open country, the experiment has got high accuracy in picking up land use data and the results can fit to the high accuracy need of land use in water and soil erosion model which indicates that the Sub-Region Classification Method are feasible to pick up the land use data in water and soil erosion model. After compare with the two results of the new method and Monitoring Classification , finds that the new method improves the accuracy of the result highly.In addition , this study try its best to make clear each type of wood , for example the tea , the new method give a viable idea for the same research .The result of the research puts the good foundation on withdrawing the water and soil erosion information further by the research.

  • 【网络出版投稿人】 浙江大学
  • 【网络出版年期】2009年 07期
  • 【分类号】F301;P237
  • 【被引频次】5
  • 【下载频次】313
节点文献中: 

本文链接的文献网络图示:

本文的引文网络