北方农业学报 ›› 2021, Vol. 49 ›› Issue (1): 127-134.doi: 10.12190/j.issn.2096-1197.2021.01.18

• 农业信息技术 • 上一篇    

基于遥感地理特征曲线的主要农作物种植面积大尺度提取

乌兰吐雅1,2, 于利峰1,2, 包珺玮1,2, 许洪滔1,2, 乌云德吉1,2, 任婷婷1,2, 赵佳乐1,2, 敦惠霞1, 于伟卓1   

  1. 1.内蒙古自治区农牧业科学院 农牧业经济与信息研究所,内蒙古 呼和浩特 010031;
    2.内蒙古自治区农业遥感工程技术研究中心,内蒙古 呼和浩特 010031
  • 收稿日期:2020-10-21 出版日期:2021-02-20 发布日期:2021-03-23
  • 通讯作者: 敦惠霞(1962—),女,编审,学士,主要从事农业信息管理与期刊出版工作。
  • 作者简介:乌兰吐雅(1970—),女,副研究员,硕士,主要从事农业遥感监测及评价的研究工作。
  • 基金资助:
    内蒙古农牧业科学院创新基金(2018年CXJJ09,2020QNJJN04)

Large scale extraction of main crops growing area based on remote sensing geographic feature curve

Wulantuya1,2, YU Lifeng1,2, BAO Junwei1,2, XU Hongtao1,2, Wuyundeji1,2, REN Tingting1,2, ZHAO Jiale1,2, DUN Huixia1, YU Weizhuo1   

  1. 1. Institute of Agricultural and Animal Husbandry Economy and Information,Inner Mongolia Academy of Agricultural and Animal Husbandry Sciences,Hohhot 010031,China;
    2. Inner Mongolia Engineering and Technology Research Center for Agricultural Remote Sensing,Hohhot 010031,China
  • Received:2020-10-21 Online:2021-02-20 Published:2021-03-23

摘要: 【目的】 研究利用遥感地理特征曲线数据快速提取内蒙古呼伦贝尔市、兴安盟的主要农作物种植面积的方法。【方法】 以2018年46期(MOD09Q1)数据为数据源,计算NDVI,利用NRF滤波方法重建时间序列数据集,并与5月NDVI 最低值、DEM数据、坡度数据、5月LSWI最高值数据集成特征曲线数据集;用CART决策树分类方法对5种农作物种植面积进行提取,并利用地面样方数据进行精度验证。【结果】 建立了小麦、油菜、玉米、大豆、水稻的标准特征曲线,并针对不同区域构建了适合该区域的农作物信息提取模型;地面样方数据验证相对精度达到77%以上。【结论】 综合考虑遥感地理特征曲线,用CART决策树分类方法大尺度快速提取主要农作物面积是可行的。

关键词: 遥感地理特征曲线, 大尺度, 农作物, 面积提取

Abstract: 【Objective】 To study the method for extracting the growing area of main crops in Hulun Buir City and Hinggan League in Inner Mongolia by using remote sensing geographic feature curve data.【Methods】 NDVI calculated by using MOD09Q1 as the dataset,46th,2018;reconstructed time series dataset by using the method of NRF filter and integrate feature curve data sets by adding NDVI minimum in May,DEM data,the slope data and LSWI maximum in May. The CART decision tree classification method was used to extract the growing area of 5 main crops,and the ground quadrat data was used to verify the accuracy.【Results】 The standard feature curve of wheat,oilseed rape,corn,soybean,and rice was built,and the crop information extraction models were constructed for the different areas;the relative accuracy of ground quadrat data verification reached more than 77%.【Conclusion】 Considering the remote sensing geographic feature curves,it was feasible to quickly extract the area of main crops on a large scale by using the CART decision tree classification method.

Key words: The remote sensing geographic feature curves, Large scale, Crop, Area extraction

中图分类号: 

  • TP79