引用本文格式刘芸,宋善海,李雪,等.基于多时相卫星影像的复杂山区烤烟种植区域提取[J].农业工程,2023,13(1):67-70.DOI:10.19998/j.cnki.2095-1795.2023.01.012.LIUYun,SONGShanhai,LIXue,etal.Extractionofflue-curedtobaccoincomplexmountainousareabasedonmulti-temporalsatelliteimages[J].AgriculturalEngineering,2023,13(1):67-70.基于多时相卫星影像的复杂山区烤烟种植区域提取刘芸1,宋善海1,李雪1,廖瑶1,李慧璇1,王伟2(1.贵州省生态气象和卫星遥感中心,贵州贵阳550002;2.贵州省特种水产工程技术中心,贵州贵阳550025)摘要:为提高卫星影像对复杂山区烤烟种植区的提取精度,以高分六号2m分辨率卫星影像作为遥感数据源,基于多时相卫星遥感影像序列,根据烤烟不同生长期的卫星影像特征,采取面向对象分类方法,分别完成3个不同时相卫星影像的烤烟种植区初步提取,再利用多时相植被指数规则初步提取烤烟种植区,最后通过交叉验证获取烤烟种植区域,并完成精度验证。结果表明,研究方法能较为准确地提取复杂山区烤烟种植区,能够保持较低的错分率(5.3%),提取精度达94.67%,较基于单一时相卫星影像提取烤烟种植区域精度(90%)明显提升。关键词:遥感;卫星影像;烤烟;面向对象分类方法中图分类号:S572文献标识码:A文章编号:2095-1795(2023)01-0067-04DOI:10.19998/j.cnki.2095-1795.2023.01.012ExtractionofFlue-curedTobaccoinComplexMountainousAreaBasedonMulti-temporalSatelliteImagesLIUYun1,SONGShanhai1,LIXue1,LIAOYao1,LIHuixuan1,WANGWei2(1.GuizhouEcologicalMeteorologicalandSatelliteRemoteSensingCenter,GuiyangGuizhou550002,China;2.GuizhouEngineeringandTechnologyCenterofSpecialFisheries,GuiyangGuizhou550025,China)Abstract:Inordertoimproveextractionaccuracyofsatelliteimagesforcomplexmountainoustobaccoplantingareas,GaofenNo.62mresolutionsatelliteimageswereusedasremotesensingdatasources.Basedonmultitemporalsatelliteremotesensingimagesequencesandsatelliteimagefeaturesofdifferentgrowthstagesoftobacco,anobject-orientedclassificationmethodwasadoptedtocompletepreliminaryextractionoftobaccoplantingareasfromthreedifferenttemporalsatelliteimages.multitemporalvegetationindexruleswereusedforpreliminaryextractionofto...