论文部分内容阅读
主要研究目的是探讨高光技术估算土壤常规元素含量的能力以及适合于这些元素的最佳光谱预处理方法。以三江源区玛多县和玉树县做为研究样区,以野外采集的149个表土层(0~30 cm)土壤样本为数据源,经过实验室化学成分测定和光谱采集,利用多种预处理方法和偏最小二乘回归法,建立五种常规元素Al、Fe、Mg、Mn、Si含量的高光谱估算模型。研究结果表明:利用偏最小二乘回归法可以估算Al、Fe、Mg含量,其最佳预处理方案分别为WT+R(小波变换+原始光谱反射率)、NWA+R(九点加权移动平均+原始光谱反射率)和WT+F(小波变换+一阶微分),模型精度分别为Rcv2=0.79,Rv2=0.76,RPD=2.01;Rcv2=0.85,Rv2=0.76,RPD=1.85;Rcv2=0.72,Rv2=0.74,RPD=2.03;该模型在此研究样区不具备估算Mn、Si含量的能力。
The main objective of the study was to investigate the ability of highlight techniques to estimate the contents of conventional elements in soils and to optimize the spectral pretreatment methods appropriate for these elements. Taking Maduo County and Yushu County of Sanjiangyuan as samples, 149 soil samples (0-30 cm) collected from field were used as data source. After laboratory chemical composition determination and spectral acquisition, Processing method and partial least-squares regression method to establish the hyperspectral estimation model of five conventional elements Al, Fe, Mg, Mn and Si. The results show that the contents of Al, Fe and Mg can be estimated by partial least squares regression. The best pretreatment schemes are WT + R (Wavelet transform + original spectral reflectance), NWA + R (nine-point weighted moving average Rcv2 = 0.85, Rv2 = 0.76, RPD = 1.85; Rcv2 = 0.72, Rv2 = 0.76, RPD = 2.01; , Rv2 = 0.74, RPD = 2.03. The model does not have the ability to estimate the contents of Mn and Si in this study area.