支持向量机在土壤镁含量高光谱估算中的应用

研究利用土壤样本实验反射光谱,分析了土壤镁(Mg)含量与土壤反射光谱的关系,比较了主成分回归分析(PCR)、偏最小二乘回归分析(PLSR)和支持向量机回归分析(SVMR)等方法,以及土壤反射光谱及其变换光谱与土壤Mg含量之间的估算模型,为土壤Mg含量高光谱估算提供依据。结果表明:PCR、PLSR、SVMR 3种建模方法在Mg含量的估算中,SVMR的估算精度相对较高,估算精度平均达到80.96%,分别比PCR和PLSR提高了6.16%、4.20%;对于不同的数学变换处理方法,一阶微分变换相对较好,估算精度平均为80.76%,分别比反射率、倒数对数变换提高了4.95%、4.61%。因此,运用土壤反射光谱一阶微分变换的SVMR进行建模,可以相对较好地估算全Mg含量,精度达84.04%。

Application of Support Vector Machine on Soil Magnesium Content Estimation Based on Hyper-Spectra

Based on the measured reflectance spectra of soil samples in the laboratory, the relationship between soil magnesium content and reflectance spectra was analyzed, and the methods of principal component regression (PCR) and partial least squares regression (PLSR) were compared with support vector machine regression (SVMR) analysis for soil magnesium content estimation . Then the estimation models between the reflectance spectra and its transforms with magnesium content were established to provide the basis for soil magnesium content estimation of hyper-spectra. The results show that SVMR has a relatively higher accuracy among the three modeling methods of PCR, PLSR and SVMR. Its average prediction accuracy reaches 80.956%. Compared with PCR and PLSR, it respectively increases by 6.162% and 4.197%. The first derivative of the reflectance spectra obtains the best outcome in the different mathematical transforms. Its average estimation accuracy is 80.759%. Compared with reflectivity and reciprocal logarithmic transforms, it improves 4.950% and 4.613%. Therefore, SVMR model of first derivative transform to estimate the total magnesium content is relatively better, which obtains the relatively higher accuracy of 84.036%.