基于多尺度连续小波分解的土壤氧化铁反演

Inversion of Soil Iron Oxide Based on Multi-Scale Continuous Wavelet Decomposition

为准确快速地预测土壤氧化铁含量信息,以禄丰恐龙谷南缘地表的土壤为研究对象,共采集135个样品,并在室内测得土壤光谱数据和氧化铁含量信息。首先,在对原始光谱进行SavitzkyGolay滤波平滑后,进行常规光谱变换和连续小波变换,并利用相关系数(CC)法对变换光谱与氧化铁含量进行相关性分析,筛选出每个尺度中通过0.01显著性检验的波长作为粗选的波长。然后,进一步利用竞争性自适应重加权(CARS)算法选择的波长作为特征波长。最后,通过遗传算法优化的支持向量机(SVR)进行建模。结果表明:连续小波变换可以提高土壤光谱反射率与氧化铁含量的相关性;通过CCCARS波长选择方法可以有效地减少建模的自变量数目;第4尺度连续小波分解构建的模型(L4-CCCARSSVR)效果最好,其建模集的决定系数R2为0.760,均方根误差ERMSE为5.236 g·kg-1,验证集的R2为0.663,ERMSE为7.798 g·kg-1,性能与四分位数间距比RPIQ达到了2.598,即模型具有很好的稳定性和预测能力。

In order to predict the content of iron oxide in soil accurately and quickly,135 soil samples are collected from the southern edge of Lufeng Dinosaur Valley,and soil spectral data and iron oxide content are measured in the laboratory.The original spectrum is smoothed by the SavitzkyGolay filter,and then conventional spectral transform and continuous wavelet transform are performed.The correlation coefficient(CC)method is used to analyze the correlation between the transform spectrum and iron oxide content.The wavelengths that pass the 0.01 significance test in each scale are selected as the coarse wavelengths,and the wavelengths selected by the competitive adaptive reweighted sampling(CARS)are further used as the characteristic wavelengths.Finally,the support vector regression(SVR)optimized by the genetic algorithm is used for modeling.The results reveal that continuous wavelet transform can improve the correlation between the soil spectral reflectance and iron oxide content.The number of independent variables for modeling can be effectively reduced by the CCCARS wavelength selection method.The model constructed by the fourthscale continuous wavelet decomposition(L4-CCCARSSVR)has the best effect.The coefficient of determination R2 and rootmeansquare error ERMSE of its calibration set is 0.760 and 5.236 g·kg-1,respectively.The R2,ERMSE and performance to interquartile range ratio RPIQ of its validation set is 0.663,7.798 g·kg-1 and 2.598,respectively,which indicates that the model has good stability and predictive ability.