北京测绘2025，Vol.39Issue(6)：855-860,6.DOI:10.19580/j.cnki.1007-3000.2025.06.017
结合多源遥感影像的河北省农用地提取
Extraction of agricultural land in Hebei Province based on multi-source remote sensing images
摘要
Abstract
In agricultural land extraction research,the use of either optical remote sensing data or synthetic aperture radar(SAR)data alone has certain limitations.This paper combined Sentinel-1 SAR data and Sentinel-2 optical remote sensing satellite data and employed the random forest algorithm to extract and classify agricultural land in Hebei Province.The results indicate that data fusion significantly improves the accuracy and robustness of agricultural land extraction.Analysis of data from Hebei Province in 2023 shows that the classification accuracy of the method incorporating Sentinel-1 SAR and Sentinel-2 data is significantly higher than that of using Sentinel-1 or Sentinel-2 images individually.The overall accuracy(OA)reaches 95.65%,and the producer's accuracy(PA)is 90.42%;the user's accuracy(UA)is 97.68%,and the Kappa coefficient is 0.903.These results demonstrate that the proposed method performs exceptionally well in agricultural land extraction.This paper provides crucial data support for precise agricultural land mapping and farmland protection,offering valuable references for future land use research.关键词
谷歌地球引擎/哨兵影像/农业用地/多源遥感融合/随机森林Key words
Google Earth Engine/sentinel image/agricultural land/multi-source remote sensing fusion/random forest分类
信息技术与安全科学引用本文复制引用
王生明,王洪杰,刘欣,郝慧迪,房馨雨..结合多源遥感影像的河北省农用地提取[J].北京测绘,2025,39(6):855-860,6.基金项目
国家重点研发计划(2021YFB3900803) （2021YFB3900803）
