首页    编委会    编辑部    历任主编    获奖信息    征稿简则    友情链接    English    旧版主页
  在线期刊
       最新录用
       当期目次
       过刊浏览

  下载专区
       版权协议
       征稿范文
       来稿登记表
       录用证明

  联系我们

投稿e-mail: jbnuns_sub@bnu.edu.cn

主编e-mail: jbnuns_eic@bnu.edu.cn

办公地点:北京师范大学 科技楼A区216,218,220室

电话:(010)58807851,58802050,58807872,58807713

通信处:北京市海淀区新街口外大街19号北京师范大学理科学报编辑部

邮政编码:100875

办公室e-mail: jbnuns@bnu.edu.cn

本网站暂不提供全文下载,如需全文下载请点击

中国知网

 

标题

图斑与变化向量分析相结合的秋粮作物遥感提取

作者

孙佩军 杨珺雯 张锦水 潘耀忠 云雅

机构

北京师范大学地表过程与资源生态国家重点实验室;北京师范大学资源学院

摘要

本文提出一种图斑与变化向量分析相结合的农作物提取方法,旨在综合面向对象空间纹理信息、归一化植被指数信息和作物变化信息,提取农作物。实验以高分一号为数据源,选取河北省衡水市中部区域提取作物,根据野外采样点数据验证方法精度。实验结果表明:该方法较高精度的提取了研究区作物分布,总体精度达到93.4%。其中主要作物玉米提取的用户精度、制图精度分别为95.4%、95.7%,棉花提取的用户精度、制图精度分别为86.8%、84.0%,其余非农作物类别用户精度高于89.9%,制图精度高于91.6%。棉花和景观树由于种植破碎,并且大部分都混作于玉米中,提取精度比玉米低。从分类结果图中可以看出,该方法基于面向对象,以地块为基本识别单元提取作物信息,消除了传统基于像元的分类存在的“椒盐现象”。研究表明,本文提出的方法能够充分利用面向对象地块信息,以农作物变化信息为依据,辅以归一化植被指数,较高精度的提取农作物。该方法的成功实施为面向对象方法在农作物提取中的研究提供了新思路,为作物信息提取提供了有效的途径。

关键词

面向对象;变化向量;农作物;分类精度

引用

孙佩军, 杨珺雯, 张锦水, 潘耀忠, 云雅. 图斑与变化向量分析相结合的秋粮作物遥感提取[J]. 北京师范大学学报(自然科学版),2015,51(Sup.1):89-94.

基金

国家自然科学基金资助项目(41301444);国家高分辨率对地观测重大专项基金资助项目(民用部分);北京高等学校“青年英才计划”资助项 目;北京市自然科学基金资助项目(8144052)

分类号

X43

DOI

10.16360/j.cnki.jbnuns.2015.s1.014

Title

Combined object-oriented and changing vector analysis for remote sensing extraction of autumn crops

Author

SUN Peijun, YANG Junwen, ZHANG Jinshui, PAN Yaozhong, YUN Ya

Affiliations

State Key Laboratory of Earth Processes and Resource Ecology, Beijing Normal University; College of Resources Science & Technology, Beijing Normal University

Abstract

In the present work we proposed a new method combining parcels based on object-oriented and changing vector analysis to extract crop, this integrated the texture information, normalized difference vegetation index (NDVI) and changing information during crop growing time to identify crop. GF-1 images were used to implement the method in central Hengshui Ctiy in Hebei province. This method was found to extract the distribution of crop accurately in general, with an overall accuracy of 93.4%. User and producer accuracy of corn, which accounted for the largest area in the study area, was 95.4% and 95.7% respectively, but the user and producer accuracy was 86.8% and 84.0% for cotton, and 89.9%, 91.6% for landscape trees. The accuracy of cotton and landscape trees was lower than corn, which were planted in small areas intercropped in the districts of corn. Thus, complexity and mixed pixel reduced accuracy. The “pepper and salt” phenomenon common in traditional pixel-based classification was reduced. This was because the present method identified the type of districts based on patches. It is concluded that the method could extract crop with high accuracy, using the information of patches based on object-oriented method, the changing information of crop and auxiliary data of NDVI. This method provides an effective way to extract crop using objected-oriented method.

Key words

object-oriented; changing vector; crop; accuracy

cite

SUN Peijun, YANG Junwen, ZHANG Jinshui, PAN Yaozhong, YUN Ya.Combined object-oriented and changing vector analysis for remote sensing extraction of autumn crops[J]. Journal of Beijing Normal University(Natural Science)),2015,51(Sup.1):89-94.

DOI

10.16360/j.cnki.jbnuns.2015.s1.014

Copyright © 2014 Journal of Beijing Normal University (Natural Science)
Designed by Mr. Sun Chumin. Email: cmsun@mail.bnu.edu.cn