欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2010, Vol. 27 ›› Issue (2): 163-169.DOI: 10.7523/j.issn.2095-6134.2010.2.003

• 论文 • 上一篇    下一篇

基于MODIS/NDVI时序数据的土地覆盖分类

刘庆凤1,2, 刘吉平1, 宋开山2   

  1. 1. 吉林师范大学旅游与地理科学学院, 四平 136000;
    2. 中国科学院东北地理与农业生态研究所, 长春 130012
  • 收稿日期:2009-06-26 修回日期:2009-11-06 发布日期:2010-03-15
  • 通讯作者: 刘吉平
  • 基金资助:

    国家重点基础研究发展计划(973)项目(2009CB421103)和吉林师范大学研究生科研创新计划项目(S09010125)资助 

Land cover classification based on MODIS/NDVI times series data

LIU Qing-Feng1,2, LIU Ji-Ping1, SONG Kai-Shan2   

  1. 1. College of Tourism & Geography Sciences, Jilin Normal University, Siping 136000, China;
    2. Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun 130012, China
  • Received:2009-06-26 Revised:2009-11-06 Published:2010-03-15

摘要:

以250m分辨率的MODIS/NDVI时间序列数据为主要数据源,通过Sacizkky-Golay滤波重建高质量NDVI时间序列数据;同时融合500m分辨率的MODIS多光谱反射率数据和90m分辨率的DEM数据.将非监督分类法和决策树法相结合,进行黑龙江流域土地覆盖分类研究.对分类结果采用已有的土地覆盖数据和高分辨率遥感影像进行精度评价,评价结果表明,利用MODIS/NDVI时间序列数据获得较高精度的土地覆盖分类结果是可行的.

关键词: 黑龙江流域, 土地覆盖, MODIS, NDVI时间序列

Abstract:

With the 250m MODIS/NDVI times series datasets as main data source we reconstructed NDVI times series datasets with higher quality by Sacizkky-Golay filter, and we merged 500m MODIS multi-spectral reflectance data with 90m DEM data. With a combination of unsupervised classification and decision-tree methods, we extracted the land cover classification of Amur-Heilong River basin. We assessed the classification accuracy by using the current land cover map and high-resolution remote sensing images. The results indicate that acquirement of more accurate land cover classification by using MODIS/NDVI times series datasets is fessible.

Key words: Amur-Heilong River basin, land cover, MODIS, NDVI time series

中图分类号: