基于GEE和Sentinel-1/2数据的夏玉米种植面积精细化识别方法
作者:
作者单位:

1.山东省气象防灾减灾重点实验室,山东 济南 250031 ;2.山东省气候中心,山东 济南 250031 ;3.长岛国家气候观象台,山东 长岛 265800 ;4.自然资源部国土空间规划研究中心,北京 100034

作者简介:

韩东枫,男,硕士,助理工程师,主要从事卫星遥感技术研究与应用,hayecer@gmail.com。

通讯作者:

中图分类号:

S127

基金项目:

新一代人工智能国家科技重大专项(2022ZD0119500);山东省自然科学基金项目(ZR2020MF130);山东省气象局气象软科学重点项目(2024SDZDIANXM01);山东省气象局科研项目(2021sdqxz03)


请扫码阅读

Recognition methods of summer maize planting areas based on GEE and Sentinel-1/2 data
Author:
Affiliation:

1.Key Laboratory for Meteorological Disaster Prevention and Mitigation of Shandong, Jinan 250031 , China ; 2.Shandong Climate Center, Jinan 250031 , China ; 3.Changdao National Climatology Observatory, Changdao 265800 , China ; 4.Research Center for Territorial Spatial Planning, Ministry of Natural Resources, Beijing 100034 , China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    作物种植面积提取方式的选取,对农作物遥感监测有重要意义。为探究夏玉米遥感识别最佳时相、夏玉米遥感识别光学时序和夏玉米遥感识别光学与星载合成孔径雷达(synthetic aperture radar,SAR)融合时序3种方案在夏玉米种植区识别的差异,选取山东商河为研究区。基于谷歌地球引擎(Google Earth Engine,GEE)云平台Sentinel-1/2数据,构建分类数据集,结合地面调查制作分类样本,采用随机森林法进行3种方案下研究区夏玉米种植区域提取,并分析各方案精度。结果表明:3种方案均能较高精度地实现夏玉米与其他作物的区分;相对于夏玉米遥感识别最佳时相方案,夏玉米遥感识别光学时序方案下夏玉米总体分类精度由83.01%提高到89.44%,Kappa系数由0.77提高到0.86;相对于夏玉米遥感识别最佳时相和夏玉米遥感识别光学时序方案,夏玉米遥感识别光学与SAR融合时序方案的总体分类精度最高,达92.51%,Kappa系数达0.89。研究表明,夏玉米遥感识别光学与SAR融合时序方案可以在较高精度下有效识别夏玉米种植区,为发育期内的农情调查管理提供参考。

    Abstract:

    The selection of methods for extracting crop planting areas is of great significance for agricultural remote sensing monitoring. To explore the differences between optimum phase scheme, time series optical data scheme and optical-SAR (synthetic aperture radar) fusion phase scheme in remote sensing recognition of summer maize planting areas, Shanghe County of Shandong Province is taken as the study area. Based on the Sentinel-1/2 data from the GEE (Google Earth Engine) cloud platform, three datasets are constructed. Combined with ground survey samples, random forest method is used to extract the summer maize planting areas in the study area using three schemes, and the accuracy of each scheme is analyzed. The result shows that all the three schemes can achieve high accuracy in distinguishing summer maize planting areas from other crops. Compared with the optimum phase scheme, the time series optical data scheme improves the overall classification accuracy of summer maize from 83.01% to 89.44%, and the Kappa coefficient increases from 0.77 to 0.86. Compared with the optimum phase scheme and time series optical data scheme, the overall classification accuracy of the optical-SAR fusion phase scheme is the highest, reaching 92.51%, and the Kappa coefficient reaches 0.89. The classification results show that the optical-SAR fusion phase scheme can effectively recognize summer maize planting areas with high accuracy, providing reference for agricultural investigation and management during the growing season.

    参考文献
    相似文献
    引证文献
引用本文

韩东枫,李峰,秦泉,等.基于GEE和Sentinel-1/2数据的夏玉米种植面积精细化识别方法[J].海洋气象学报,2024,44(3):122-132.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
文章历史
  • 收稿日期:2024-01-28
  • 最后修改日期:2024-04-18
  • 录用日期:
  • 在线发布日期: 2024-09-04
  • 出版日期:

“海雾专题”征稿通知

关闭