论文部分内容阅读
土壤光谱反射特性是土壤遥感的物理基础。通过野外调查采样、土壤盐分实验分析与土壤高光谱数据采集,对土壤高光谱数据一阶和二阶导数微分变换处理,分析土壤样品的光谱特征,建立土壤光谱和土壤盐分含量间的相关关系,对研究区盐渍化土壤含盐量进行定量反演。研究结果表明:1)从土壤光谱反射率的形态特征来看,土壤的光谱反射率曲线总体上变化较为平缓,光谱特征形态较为相似,且基本平行。2)研究区土壤光谱反射率曲线的形状大致可由300~600nm、600~800nm、800~1000nm、1000~1400nm、1400~1900nm、1900~2100nm、2100~2500nm七个折线段和560nm、900nm、1400nm、1900nm、2200nm五个特征吸收点来控制。3)利用光谱反射率一阶导数微分的盐渍化土壤含盐量多元线性回归预测模型的预测效果均优于利用反射率原型和反射率二阶导数微分,其中氯化物-硫酸盐型RMSE=0.33,硫酸盐型RMSE=0.31,硫酸盐-氯化物型RMSE=0.22。
Soil spectral reflectance is the physical basis of soil remote sensing. Through the field investigation and sampling, soil salinization experiment and soil hyperspectral data acquisition, the first and second derivative differential transformation of soil hyperspectral data were processed, the spectral characteristics of soil samples were analyzed, and the correlation between soil spectra and soil salinity content was established. The salinity soil salinization in the study area was quantitatively retrieved. The results show that: 1) The spectral reflectance curves of soils generally change more slowly and the spectral features are similar and basically parallel, according to the morphological characteristics of soil spectral reflectance. 2) The spectral reflectance curve of soils in the study area can be roughly composed of seven broken line segments of 300-600 nm, 600-800 nm, 800-1000 nm, 1000-1400 nm, 1400-1900 nm, 1900-2100 nm and 2100-2500 nm, and 560 nm, 900 nm, 1400 nm , 1900nm, 2200nm five characteristics of absorption point to control. 3) Prediction results of multivariate linear regression model of salinity soil salinity using spectral derivative first derivative differential are better than those using second derivative of reflectivity and reflectance model. Among them, chloride-sulfate type RMSE = 0.33, Sulfate RMSE = 0.31, Sulfate-Chloride RMSE = 0.22.