期刊文献+

基于多特征优选的平和县蜜柚园遥感提取与扩张分析 被引量:2

Spatio-temporal Change Analysis of Pomelo Orchards in Pinghe County Based on Multi-feature Optimization
下载PDF
导出
摘要 本文基于谷歌地球引擎(google earth engine,GEE)平台中2020年Sentinel-2数据和数字高程模型(digital elevation model,DEM)数据,在递归特征消除的随机森林(random forest-recursive feature elimination,RF-RFE)特征选择算法基础上结合随机森林(random forest,RF)分类算法,实现了平和县蜜柚园的空间分布制图。根据蜜柚园扩张的先验知识,使用2020年蜜柚园的提取结果对历史Landsat数据进行掩膜,依次获得1990、2000、2010年平和县蜜柚园的空间分布,在此基础上对30年以来平和县蜜柚园的变化进行分析。结果表明:基于GEE平台,使用RF-RFE特征选择后的多特征分类方法可以快速、准确地提取平和县蜜柚园的空间分布。蜜柚园的生产精度和用户精度分别为89.83%和90.59%;平和县蜜柚园面积从1990年的10.5 km^(2)增加到2020年632 km^(2),增加的蜜柚园主要来源于有林地,其次为耕地和香蕉园。本研究可为平和县蜜柚产业的健康发展提供决策支持,同时可为南方地区的果园提取提供技术参考。 Based on the Sentinel-2 data and digital elevation model data in google earth engine platform in 2020,and based on random forest-recursive feature elimination algorithm and random forest classification algorithm,the spatial distribution mapping of pomelo orchard in Pinghe County was achieved.According to the priori-knowledge of pomelo orchards expansion,the extraction results of pomelo orchards in 2020 were used to mask the data of Landsat in 1990,2000 and 2010,and the spatial distribution of pomelo orchards in Pinghe County in 1990,2000 and 2010 was obtained.On this basis,the dynamic changes of pomelo orchards in Pinghe County in the past 30 years were analyzed.The results show that the multi-feature classification method based on GEE platform using RF-RFE feature selection can quickly and accurately extract the spatial distribution of pomelo orchards in Pinghe County.The classification results for pomelo orchards showed that the overall accuracy and Kappa coefficient were 92.09%and 0.87,respectively,and the pomelo orchards producer's accuracy and user's accuracy were 89.83%and 90.59%,respectively.The area of pomelo orchards in Pinghe County increased from 10.5 km^(2) in 1990 to 632 km^(2) in 2020,the increase of pomelo orchards mainly came from forest land,followed by cultivated land and banana garden.This study can provide decision support for the healthy development of honey pomelo industry in Pinghe County.At the same time,it can provide method guidance for orchard extraction in southern region.
作者 刘雪萍 周小成 崔雅君 肖祥希 LIU Xueping;ZHOU Xiaocheng;CUI Yajun;XIAO Xiangxi(Key Laboratory of Spatial Data Mining and Information Sharing of Ministry of Education,National&Local Joint Engineering Research Center of Satellite Geospatial Information Technology,Fuzhou University,Fuzhou 350108,China;Fujian Academy of Forestry Sciences,Fuzhou 350012,China)
出处 《贵州大学学报(自然科学版)》 2022年第3期52-61,92,共11页 Journal of Guizhou University:Natural Sciences
基金 福建省林业局重点科技攻关资助项目(2021FKJ01) 中国科学院战略性先导科技专项资助项目(XDA23100504)。
关键词 蜜柚园 Sentinel-2 随机森林算法 RF-RFE 平和县 pomelo orchards Sentinel-2 random forest algorithm RF-RFE Pinghe County
  • 相关文献

参考文献20

二级参考文献235

共引文献501

同被引文献37

引证文献2

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部