测绘通报 ›› 2017, Vol. 0 ›› Issue (2): 19-24.doi: 10.13474/j.cnki.11-2246.2017.0041

• 学术研究 • 上一篇    下一篇

高分辨率遥感影像下沿海地区地表覆盖信息的提取

周星宇1, 张继贤2, 高绵新3, 桑会勇2, 翟亮2   

  1. 1. 辽宁工程技术大学, 辽宁 阜新 123000;
    2. 中国测绘科学研究院, 北京 100830;
    3. 广东省国土资源测绘院, 广东 广州 510500
  • 收稿日期:2016-08-29 修回日期:2016-11-25 出版日期:2017-02-25 发布日期:2017-03-01
  • 通讯作者: 翟亮。E-mail:zhailiang@casm.ac.cn E-mail:zhailiang@casm.ac.cn
  • 作者简介:周星宇(1991-),女,硕士,研究方向为地理国情监测、土地利用变化。E-mail:xiaoyu961026631@163.com
  • 基金资助:

    专题性地理国情监测(B1605);国家测绘地理信息局青年学术和技术带头人科研计划(E1604);中国测绘科学研究院基本科研业务费(7771622);京津冀地区基础国情综合分析(E1610)

Land Cover Information Extraction Based on High-Resolution Remote Sensing Image in Coastal Areas

ZHOU Xingyu1, ZHANG Jixian2, GAO Mianxin3, SANG Huiyong2, ZHAI Liang2   

  1. 1. Liaoning Technical University, Fuxin 123000, China;
    2. Chinese Academy of Surveying and Mapping, Beijing 100830, China;
    3. Survey and Mapping Institute Lands and Resource Department of Guangdong Province, Guangzhou 510500, China
  • Received:2016-08-29 Revised:2016-11-25 Online:2017-02-25 Published:2017-03-01

摘要:

沿海地区地表覆盖信息是全国地理国情普查的重要内容,遥感影像分类技术为沿海地区地表覆盖信息提供了一种重要方法。本文基于GF-1高分辨率遥感影像,建立了沿海地区地表覆盖分类系统,采用中国测绘科学研究院自主研发的面向对象GLC决策树分类方法和软件进行了地表覆盖分类。通过对某试验区进行分类试验,并结合该区地表覆盖标准分类图进行精度评价,验证了基于高分辨率影像,面向对象GLC决策树分类方法在沿海地区地表覆盖信息提取上的有效性及优越性,其总体分类精度和Kappa系数分别为87.201 8%、0.840 6,均高于SVM分类法。最后提出基于高分辨率遥感影像的沿海地区地表覆盖信息提取流程。

关键词: 沿海地区地表覆盖, 面向对象, GLC树, GF-1, SVM分类法, 提取流程

Abstract:

Remote sensing image classification provides an important method for the extraction of land cover information in coastal areas which is essential part of the national general survey of geographic conditions. This paper establishes the land cover classification system in coastal areas and then utilizes the classification method based on object-oriented GLC decision tree developed by Chinese Academy of Surveying and Mapping to extract the land cover information in coastal areas on the bases of a GF-1 high-resolution remote sensing image. This paper conducts classification experiment by choosing an area and compares the results with the reference classification image which verifies the validity and superiority of the proposed method. Its overall accuracy and Kappa coefficient are 87.201 8%,0.840 6 separately which are both higher than SVM. At the end of this thesis, the extraction process flow of land cover information in coastal areas based on the high-resolution remote sensing image is summarized.

Key words: land cover in coastal areas, object-oriented, GLC decision tree, GF-1, SVM, extraction process flow

中图分类号: