基于环境星的混合像元分解水稻面积提取  

Extraction of Paddy Planting Areas by Using Mixed Pixel Decomposition Based on HJ-1A/B Satellite Images

作　　者：陈树辉[1] 李杨[2] 曾凡君[1,2] 薛春燕[2] 刘翔[1] 

机构地区：[1]北京东方泰坦科技股份有限公司,北京100083 [2]南京林业大学森林资源与环境学院,江苏南京210037

出　　处：《安徽农业科学》2011年第10期6104-6106,共3页Journal of Anhui Agricultural Sciences

基　　金：国家"863"计划地球观测与导航技术领域国家统计遥感业务系统关键技术研究与应用专题项目(2006AA120107)

摘　　要：以河南省息县水稻种植面积提取为例,选取线性光谱混合模型对环境小卫星数据进行分类,并计算出水稻种植面积,将其结果与高分辨率ALOS数据进行位置精度计算,得到平均精度达87.89%。同时,与决策树分类方法和神经元网络分类等方法进行对比,混合像元分解方法总量精度显著提高。表明针对环境小卫星的混合像元分解方法可以提高水稻种植面积的提取精度。Taking the case of the extraction of paddy planting areas in Xi County of Henan Province,Linear Spectral Mixture Model（LSMM） was selected to classify and calculate paddy planting areas,moreover,location accuracy calculation was carried out for its result and high resolution ALOS data,thereby,average precision reached 87.89%.Meanwhile,comparing with decision tree classification method and neural network classification,total precision of mixed pixel decomposition method improved remarkably.The results indicated that the extraction accuracy of paddy planting areas can be improved by using mixed pixel decomposition method of HJ-1A/B satellite.

关 键 词：环境小卫星 混合像元分解 端元 线性光谱混合模型 水稻种植面积 

分 类 号：S127[农业科学—农业基础科学]