北方农业学报 ›› 2017, Vol. 45 ›› Issue (2): 109-109.

• 农业经济·农业信息技术 • 上一篇    下一篇

基于Landsat 8 OLI影像的大兴安岭西麓春小麦识别方法的比较研究

乌兰;乌兰吐雅;包珺玮   

  1. 内蒙古农牧业科学院农牧业经济与信息研究所,内蒙古呼和浩特010031
  • 出版日期:2017-04-20 发布日期:2017-04-20
  • 通讯作者: 乌兰
  • 作者简介:乌兰(1985-),女,助理研究员,博士,主要从事农业和草原遥感及GIS应用研究工作。
  • 基金资助:
    内蒙古农牧业科学院创新基金项目(2015CXJJN05);内蒙古财政厅项目;内蒙古农牧业科学院青年基金项目(2017QNJJN10)

Study on comparison of identification methods for spring wheat at Great Khingan western slope based on Landsat 80LI

Wulan, Wulantuya, BAO Junwei (Institute of Agricultural Economic and Information, Inner Mongolia Academy of Agricultural & Animal Husbandry Sciences, Hohhot 010031 ,China)   

  • Online:2017-04-20 Published:2017-04-20

摘要: 采用Landsat 8 OLI影像数据和野外调查数据,结合物候学、目视解译、非监督分类ISODATA法和监督分类最大似然法,以及面向对象分类法,对大兴安岭西麓农场的春小麦识别进行了初步探索。结果表明:最大似然法、ISODATA法和面向对象分类法的分类总体精度分别为53.33%、54.65%和71.26%;三种方法的春小麦用户精度分别为60.00%、61.29%和83.33%。在大兴安岭西麓苗期的春小麦识别中,面向对象分类法优于/SODATA法和最大似然法,为该区域春小麦识别提供思路。

Abstract: Using Landsat 80LI image and field survey data, combined with visual interpretation, ISODATA classification of unsupervised classification, maximum likelihood classification of supervised classification and object-oriented image classification, a preliminary study was conducted on spring wheat identification at farmland in Great Khingan western slope. The results showed the whole accuracy of classification accuracy in maximum likelihood, ISODATA and object-oriented image classification were 53.33% ,54.65%;~ 71.26% respectively; user accuracy of spring wheat in three methods were 60.00%, 61.29% and 83.33% respectively. The object-oriented image classification was better than ISODATA and maximum likelihood classification, and it would provide ideas for spring wheat identification in these areas.

中图分类号: 

  • S127