摘要
不同植被之间、植被不同季节的光谱特征是进行城市植被遥感动态监测的重要内容。本研究基于实测的7种上海城市典型园林植被在夏季和冬季的光谱数据,分析了这些植被不同季相下的光谱变化特征,包括同一植被叶片与冠层的光谱差异。在此基础上,借助连续统去除法结合不同季相植被光谱差异分析了不同植被的可区分光谱波段。结果表明:供测植被在冬夏季的光谱特征有一些差异,冬季植被的反射率明显低于夏季,可见光波段与近红外波段分别低约60%和40%。同种植被的冠层和叶片光谱在近红外波段的反射率不同,主要取决于该植被的盖度、冠层结构。夏季乔木的光谱反射率在全波段普遍高于灌木,而在冬季正好相反。供测植被中,夹竹桃在夏季和冬季,全波段反射率明显高于其他植被达40%~60%。基于连续统去除法的光谱变换处理,能够有效地放大植被光谱间的差异,由此获得了植被间的可分波段,原本在可见光波段可分的区域通过连续统去除变换后,可分区域面积扩大。
Spectral characteristics of seven typical landscapes in Shanghai were studied to find out the differentiable spectral bands for dynamic monitoring of urban vegetations.The results show that the differences of spectral characteristics of vegetations between winter and summer are not very distinctive.However,the reflectance of vegetation in winter is significantly lower than that in summer,which is about60%and 40%lower in the visible band and near infrared band respectively.The reflectance of the canopy and leaf spectra of the same vegetation in the near infrared band is different,depending on the coverage and canopy structure of the vegetation.In summer,the spectral reflectance of trees is generally higher than that of shrubs,and in winter,it is the opposite.The reflectance of Nerium indicumis significantly higher than other vegetations both in summer and winter.By means of continuum removal,the spectral differences between vegetations can be effectively amplified,and the bands for separating vegetations can be obtained.
作者
项巧巧
申广荣
吴裕
康宏樟
朱赟
XIANG Qiao-qiao;SHEN Guang-rong;WU Yu;KANG Hong-zhang;ZHU Yun(School of Agriculture and Biology,Shanghai Jiaotong University,Shanghai 200240,China;Key Laboratory of Urban Agriculture,Ministry of Agriculture,Shanghai 200240,China;Shanghai Urban Forest Research Station,State Forestry Administration,Shanghai 200240,China)
出处
《上海交通大学学报(农业科学版)》
2018年第5期14-21,共8页
Journal of Shanghai Jiaotong University(Agricultural Science)
基金
高分国土资源遥感应用示范系统(一期)项目(04-Y30B01-9001-12/15)
上海交大农工交叉项目(Agri-X2015004)
关键词
植被
冠层
季相
特征光谱
连续统去除法
vegetation
canopy
seasonal phase
characteristic spectrum
continuum removal