摘要
以蒙古黄芪种苗为试材,采用高光谱和多元线性回归的方法,研究了栽培和仿野生对蒙古黄芪黄芪甲苷、毛蕊异黄酮葡萄糖苷含量估测方法的影响,以期为实践中利用高光谱快速、准确估测栽培和仿野生蒙古黄芪的药效成分含量提供参考依据。结果表明:栽培和仿野生蒙古黄芪药效成分含量的原始高光谱数据所建立模型的稳定性和拟合度均是最好。估测黄芪甲苷含量模型的RMSE分别是0.0045、0.0085;R^(2)分别是0.761、0.879;所建模型分别是y=0.0054+0.0016x_(1)+0.0002x_(2)+0.0007x_(3)(R^(2)=0.903)、y=-0.1223+0.0040x_(1)+0.0001x_(2)-0.0009x_(3)(R^(2)=0.904)。估测毛蕊异黄酮葡萄糖苷含量模型的RMSE分别是0.0017、0.0040;R^(2)分别是0.860、0.868;所建模型分别为y=0.0731-0.0080x_(1)-0.0004x_(2)+4×10-6x_(3)(R^(2)=0.891)、y=0.0842-0.0007x_(1)-0.0001x_(2)+0.0002x_(3)(R^(2)=0.883)。
Taking A.mongholicus seedlings as experimental materials,the effects of cultivated and imitating wild on the estimation methods of astragaloside and calycosin-7-glucoside content in A.mongholicus were studied by using hyperspectral and multiple linear regression methods,in order to provide reference for the rapid and accurate estimation of the medicinal composition content of cultivated and imitating wild A.mongholicus in practice.The results showed that the stability and the fitting degree of the models established by the raw hyperspectral data of medicinal composition content of cultivation and imitating wild A.mongholicus were the best.The RMSE of the models to estimate the astragaloside content were 0.0045,0.0085;the R^(2) were 0.761,0.879;the models were y=0.0054+0.0016x_(1)+0.0002x_(2)+0.0007x_(3)(R^(2)=0.903),y=-0.1223+0.0040x_(1)+0.0001x_(2)-0.0009x_(3)(R^(2)=0.904),respectively.The RMSE of the models to estimate the calycosin-7-glucoside content were 0.0017,0.0040;the R^(2) were 0.860,0.868;the models were y=0.0731-0.0080x_(1)-0.0004x_(2)+4×10-6x_(3)(R^(2)=0.891),y=0.0842-0.0007x_(1)-0.0001x_(2)+0.0002x_(3)(R^(2)=0.883),respectively.
作者
刘杰
郭嘉华
赵鹏
郭宁
邢颖
段天凤
LIU Jie;GUO Jiahua;ZHAO Peng;GUO Ning;XING Ying;DUAN Tianfeng(Ecological Environment College,Baotou Teachers′College,Baotou,Inner Mongolia 014030;School of Educational Sciences,Baotou Teachers′College,Baotou,Inner Mongolia 014030)
出处
《北方园艺》
CAS
北大核心
2024年第2期100-108,共9页
Northern Horticulture
基金
中央引导地方科技发展资金资助项目(2020ZY0025)
内蒙古自然科学基金资助项目(2020MS03071)
包头师范学院自然科学类青年科研资助项目(BSYKJ2023-ZQ10)。