基于高光谱的南疆盆地苹果光合有效辐射估算模型

摘要

[目的]通过将原始光谱数据经过不同的数据变换方武,分析其与苹果冠层PAR的相关关系,建立基于高光谱的南疆盆地苹果冠层PAR的估测模型,旨在为快速、精确、无损伤、大范围的适时、动态监测植被PAR提供有效途径.[方法]基于原始光谱数据的不同数据变换方式,采用相关性分析和逐步回归分析方法.[结果]不同数据变换后的冠层光谱反射率与苹果冠层PAR具有较好的相关性,微分变换后的相关性较原始相关性有所提升.所建模型经过精度评价发现,原始光谱数据经倒数一阶微分变换后估测模型拟合度最高,一阶微分、对数一阶微分次之.[结论]不同数据变换方式后的光谱数据与塔里木盆地苹果冠层PAR有一定的相关性,可以用微分、对数微分、倒数微分变换后的数据建立较理想的塔里木盆地苹果冠层PAR的估测模型.%[Objective]Through the different data transformation of original spectral data,to analyze its correlativity between apple canopy PAR and establish the hyperspectral model of apple canopy PAR in the southern Xinjiang based on hyperspectral,the present paper aims to provide effective ways for a fast and accurate,no-damage,and wide range of timely,dynamic monitoring of vegetation PAR.[Method]Based on the different data transformation methods of the original spectral data,adapt correlation analysis and stepwise regression analysis.[Result] The canopy spectral reflectance with different transferred data has great correlations between apple canopy PAR,the correlation after differential transformation was higher than that of the original.After the accuracy evaluation,it is found in model that the original spectral data after the reciprocal first order differential transformation,the estimation model has the highest degree of fitting,and the first order differential,logarithmic first order differential are following.[Conclusion] The spectral data with different data transformation have certain correlation with apple canopy in Tarim Basin,and it can use the data with differential,logarithmic differential,reciprocal differential transformation to establish estimated model of tarim basin apple canopy PAR.