摘要
为克服光谱估测中的不确定性和提高光谱估测精度,本文利用灰色系统理论和模糊理论建立土壤有机质高光谱估测模型。基于山东省济南市章丘区和济阳区的121个土壤样本数据,首先对土壤光谱数据进行光谱变换,根据极大相关性原则选取光谱估测因子;然后,利用区间灰数的广义灰度对建模样本和检验样本的估测因子进行修正,以提高相关性。最后,利用模糊识别理论建立土壤有机质高光谱自反馈模糊估测模型,并通过调整模糊分类数进行模型优化。结果表明,利用区间灰数的广义灰度可有效提高土壤有机质含量与估测因子的相关性,所建估测模型精度和检验精度均显著提高,其中20个检验样本的决定系数为R^(2)=0.9408,平均相对误差为6.9717%。研究表明本文所建立的土壤有机质高光谱自反馈灰色模糊估测模型是可行有效的。
To overcome the uncertainty in spectral estimation and improve the accuracy of spectral estimation,a hyper-spectral estimation model of soil organic matter is established in this paper by using grey system theory and fuzzy theory.Based on 121 soil samples from Zhangqiu and Jiyang districts of Jinan City,Shandong Province,the spectral data are firstly transformed and the spectral estimation factors are selected according to the principle of great correlation;then,the estimation factors of the modeling samples and the test samples are corrected by using the generalized greyness of the interval grey number to improve the correlation.Finally,the fuzzy estimation model with self-feedback of soil organic matter based on hyper-spectral is established by using the fuzzy recognition theory,and the model is optimized by adjusting the fuzzy classification number.The results show that the correlation between soil organic matter content and estimation factors can be effectively improved by using the generalized greyness of interval grey number,and the accuracy of the built estimation model and the test accuracy are significantly improved,among which the determination coefficient of 20 test samples is R^(2)=0.9408,and the average relative error is 6.9717%.The study indicates that the grey fuzzy estimation model with self-feedback of soil organic matter using hyper-spectral data developed in this paper is feasible and effective.
作者
于锦涛
李西灿
曹双
刘法军
YU Jin-tao;LI Xi-can;CAO Shuang;LIU Fa-jun(School of Information Science and Engineering/Shandong Agricultural University,Tai’an 271018,China;The Fifth Geological Brigade of Shandong Geological and Mineral Resources Exploration and Development Bureau,Tai’an 271000,China)
出处
《山东农业大学学报（自然科学版）》
北大核心
2023年第4期495-499,共5页
Journal of Shandong Agricultural University：Natural Science Edition
基金
泰安市科技创新发展项目(2021NS090)
山东省自然科学基金项目(ZR2022QG037)。
关键词
土壤有机质
高光谱遥感
估测模型
Soil organic matter
Hyper-spectral remote sensing
stimation model
