详细信息
基于多时相遥感影像的作物种植信息提取 ( EI收录) 被引量:77
Crops planting information extraction based on multi-temporal remote sensing images
文献类型:期刊文献
中文题名:基于多时相遥感影像的作物种植信息提取
英文题名:Crops planting information extraction based on multi-temporal remote sensing images
作者:张健康[1,2] 程彦培[1] 张发旺[1] 岳德鹏[2] 郭晓晓[2] 董华[1] 王计平[3] 唐宏才[1]
第一作者:张健康
通信作者:Cheng, Y.
机构:[1]中国地质科学院水文地质环境地质研究所;[2]北京林业大学省部共建森林资源培育与保护教育部重点实验室;[3]中国林业科学研究院森林生态环境与保护研究所
年份:2012
卷号:28
期号:2
起止页码:134-141
中文期刊名:农业工程学报
外文期刊名:Transactions of the Chinese Society of Agricultural Engineering
收录:CSTPCD;;EI(收录号:20120314688328);Scopus(收录号:2-s2.0-84862929024);北大核心:【北大核心2011】;CSCD:【CSCD2011_2012】;
基金:国土资源部公益性行业专项华北平原典型地区水资源约束下的土地合理利用与管制技术研究(200811072)
语种:中文
中文关键词:遥感;影像分析;信息技术;MODIS;EVI;决策树分类;信息提取
外文关键词:remote sensing; image analysis; information technology; MODIS; EVI; decision tree classification; information extraction
分类号:S127
摘要:为了快速、准确地在遥感影像上对作物种植信息进行提取,该研究运用多时相的TM/ETM+遥感影像数据和13幅时间序列的MODISEVI遥感影像数据,采取基于生态分类法的监督分类与决策树分类相结合的人机交互解译方法,建立决策树识别模型,对黑龙港地区的主要作物进行遥感解译,总体分类精度达到了91.3%,与单纯对TM影像进行监督分类相比,棉花、玉米、小麦、蔬菜4类作物的相对误差的绝对值分别降低了1.3%、20.5%、2.0%、13.8%。结果表明该方法的分类精度高,能较好的反映作物的分布状况,可为该地区主要作物种植结构调整提供科学依据,还可为其他区域尺度作物分布信息的提取提供参考。
The multi-temporal remote sensing data were used to extract crops planting information quickly and accurately from TM/ETM+ remote sensing images and thirteen MODIS time series remote sensing images,together with the supervised classification and decision tree classification system to interpret major crops in the Heilonggang area.Overall,classification accuracy was up to 91.3%.Compared with one simple supervised classification of TM images,the relative errors of cotton,maize,wheat and vegetables reduced by 1.3%,20.5%,2.0% and 13.8% respectively.It proved that this method has high accuracy and it is a good index for the crop planting distribution.The data can provide important scientific information for the adjustment of the major crops planting structure in Heilonggang area and application references for crops classification and crop planting extraction in other area.
参考文献:
正在载入数据...