肖志云 徐新宇

摘要 針对土默川平原地区的土壤盐分含量提出了偏最小二乘与随机森林相结合（RF-PLSR、PLSR-RF）对土壤盐分含量进行预测的回归反演模型。该研究共采集45份土壤样本，随机选取35份为建模集，10份为验证集。试验首先对采集到的高光谱土壤图像进行分割处理提取出土壤在400～1 000 nm的原始反射光谱，其次对原始反射光谱进行4种光谱变换（一阶微分、多元散射校正的一阶微分、SG平滑去噪的一阶微分、对数的一阶微分），并与土壤的实测盐分量进行相关性分析（CA），利用相关系数选取敏感波段，最后建立偏最小二乘与随机森林结合的回归反演模型。结果表明，与偏最小二乘回归、随机森林回归单独建模相比，2种模型结合后的预测精度有明显的改善。光谱经过对数的一阶微分变换建立的PLSR-RF反演模型更为明显，其建模集决定系数Rc2为0.852，均方根误差RMSEc为0.102 g/kg，相对分析误差RPDc为2.600，验证集决定系数Rv2为0.941，均方根误差RMSEv为0.049 g/kg，相对分析误差RPDv为4.117。

关键词 高光谱;土壤盐含量;光谱变换;偏最小二乘回归;随机森林回归

中图分类号 TP391.4;TP79文献标识码 A

文章编号 0517-6611（2021）08-0010-06

doi：10.3969/j.issn.0517-6611.2021.08.004

开放科学（资源服务）标识码（OSID）：

Research on Inversion of Soil Salt Content Based on Partial Least Squares Combined with Random Forest

XIAO Zhi-yun1，2，XU Xin-yu1，2 （1.College of Electric Power，Inner Mongolia University of Technology，Huhhot，Inner Mongolia 010080;2.Inner Mongolia Key Laboratory of Mechatronic Control，Huhhot，Inner Mongolia 010051）

Abstract Aiming at the soil salt content in the Tumochuan Plain，a regression inversion model combining partial least squares and random forest （RF-PLSR，PLSR-RF） to predict soil salt content was proposed.A total of 45 soil samples were collected in the study，35 of which were randomly selected as the modeling set and 10 of which were randomly selected as the verification set. The experiment first performed segmentation processing on the collected hyperspectral image of the soil to extract the original reflection spectrum of the soil at 400-1 000 nm，and then performed 4 kinds of spectral transformations on the original reflection spectrum （first-order differential，first-order differential of multiple scattering correction，SG smoothing Denoising first-order differential and logarithmic first-order differential）. And it performed correlation analysis （CA） with the measured salt content of the soil，utilized the correlation coefficient to select the sensitive band，and finally established a regression model combining partial least squares and random forest. Compared with partial least square regression and random forest regression，the prediction accuracy of the combination of the two models was significantly improved. The PLSR-RF inversion model that established by the first-order differential transformation of the spectrum was more obvious. Its modeling set determination coefficient Rc2 was 0.852，the root mean square error RMSEc was 0.102 g/kg，and the relative analysis error RPDc was 2.600. The set determination coefficient Rv2 was 0.941，the root mean square error RMSEv was 0.049 g/kg，and the relative analysis error RPDv was 4.117.

Key words Hyperspectral; Soil salt content; Spectral transformation;Partial least squares regression; Random forest regression

土壤盐碱化是目前世界面临的最主要的环境问题之一，直接影响着农业的可持续发展，而土默川平原地区是内蒙古主要的粮食生产基地之一，由于特定的水文地质条件、不合理的耕作和灌溉系统，该地区出现了大面积的盐碱地[1]。土地盐渍化问题变得越来越严重，这严重影响了该地区农牧民的收入和农业生产[2]。……