首页 | 本学科首页   官方微博 | 高级检索  
     

基于多因子对象的高空间分辨率遥感影像道路提取
引用本文:施益强,朱晓铃,蔺方. 基于多因子对象的高空间分辨率遥感影像道路提取[J]. 厦门水产学院学报, 2010, 0(4): 312-316
作者姓名:施益强  朱晓铃  蔺方
作者单位:[1]集美大学影像信息工程技术研究中心,福建厦门361021 [2]集美大学理学院,福建厦门361021 [3]厦门理工学院空间信息技术研究所,福建厦门361024
基金项目:福建省高校服务海西建设重点项目(闽教高[2009]8号);福建省教育厅科技项目(JA07188)
摘    要:
利用面向对象思想,综合应用光谱值、光滑度、紧凑度及长宽比等因子,分割道路对象,构建规则知识库,探讨一种基于多因子对象的高空间分辨率遥感影像道路提取方法,并以厦门市局部区域的QuickBird影像为例进行实证.结果表明:影像分割尺度为75时,道路对象被较完整分割;与传统基于单个像元光谱信息的监督分类法相比,该方法的提取精度较高.

关 键 词:高空间分辨率  遥感影像  多因子对象  道路提取

Road Extraction in High Spatial Resolution Remote Sensing Image Based on Multi-Factor Objects
SHI Yi-qiang,ZHU Xiao-ling,LIN Fang. Road Extraction in High Spatial Resolution Remote Sensing Image Based on Multi-Factor Objects[J]. , 2010, 0(4): 312-316
Authors:SHI Yi-qiang  ZHU Xiao-ling  LIN Fang
Affiliation:1. Research Center of Image Information Engineering and Technology,Jimei University,Xiamen 361021,China; 2. School of Sciences,Jimei University,Xiamen 361021,China; 3. Institute of Spatial Information Technology, Xiamen University of Technology,Xiamen 361024,China)
Abstract:
Using object-oriented idea,considering such factors as spectrum value,smooth,compaction and aspect ratio,segmenting the road object,building a library of rule learning,this paper proposed a method of road extraction in high spatial resolution remote sensing image based on multi-factor objects. Partial QuickBird image of Xiamen City was choosen as experimental data to verify the proposed method. The experiment result showed that the road objects were fully segmented when the segmentation scale was 75,and that the extraction precision by this method was higher than that by the traditional supervised classification based on spectrum information of single pixel.
Keywords:high spatial resolution remote sensing image multi-factor objects road extraction
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号