基于MODIS 遥感数据的混合像元分解技术研究和应用Pixel Unmixing .PDF

312 Vol.31 No.2 2008 4 Journal of Nanjing Institute of Meteorology Apr.2007 , , , . MO IS [J] . , 2008, 31( 2) : 145- 150. MODIS 1 1 2 1 郑有飞 , 范旻昊, 张雪芬, 吴荣军 ( 1. , 210044;2. , 450003) :使用郑州市MO IS( Moderate-Resolution Imaging Spectroradiometer) 遥感数据, 运用线性 混合模型, 对MO IS 遥感数据进行混合像元分解技术研究探讨了MO IS 遥感数据的 处理 线性光谱分解模型图像端元组分反射率的求取方法把结果与分辨率较高的Landsat ET M+ 图 像分类结果进行对比,并根据得到的均方根误差( RMS; Root Mean Square) 进行分析表明, 利用这 种像元分解方法得到的结果较为理想,MO IS 数据可以有效地应用于遥感动态监测和土地覆盖分 类研究 :MO IS 数据; 混合像元;遥感; 线性模型 :P407 :A : 1000-2022( 2008)02-0145-06 Pixel Unmixing Technology of MODIS Remote Sensing Data 1 1 2 1 ZHENG You-fei , FAN Min-hao ,ZHANG Xue-fen ,WU Rong-jun ( 1.School of Environmental Science and Engineering, NU IST, Nanjing 210044, China; 2. Henan Research Institute of M eteorological Sciences, Zhengzhou 450003, China) Abst act: MO IS( Moderate-Resolution Imaging Spectroradiometer) remote sensing data have higher radiometric sensitivity,but its lower spatial resolution always causes the pixels impurity, normal classification methods of land cover can not solve this problem. In order to achieve the classification of land cover a line spectral mixture model was used in MO IS data pixel unmixing of Zhengzhou area. The preprocessing of MO IS data, line spectral mixture model( LSMM) , and methods for solving image end-members reflectance were also discussed.T he classification result was compared with the class map derived from Landsat ET M+ data. T he RMS shows that the pixel unmixing method has good results and the MO IS data can be effectively applied to remote sensing dynamic monitoring,and land cover c