您好，欢迎访问云南省农业科学院 机构知识库！

Rapid and Accurate Authentication of Porcini Mushroom Species Using Fourier Transform Near-Infrared Spectra Combined with Machine Learning and Chemometrics

文献类型： 外文期刊

作者： Liu, Hong 1 ; Liu, Honggao 3 ; Li, Jieqing 1 ; Wang, Yuanzhong 2 ;

作者机构： 1.Yunnan Agr Univ, Coll Agron & Biotechnol, Kunming 650201, Peoples R China

2.Yunnan Acad Agr Sci, Med Plants Res Inst, Kunming 650200, Peoples R China

3.Zhaotong Univ, Yunnan Key Lab Gastrodia & Fungi Symbiot Biol, Zhaotong 657000, Yunnan, Peoples R China

期刊名称：ACS OMEGA （ 影响因子：4.1； 五年影响因子：4.0 ）

ISSN： 2470-1343

年卷期： 2023 年 8 卷 22 期

页码：

收录情况： SCI

摘要： Porcini mushrooms have high nutritional value and greatpotential,but different species are easily confused, so it is essential to identifythem rapidly and precisely. The diversity of nutrients in stipe andcap will lead to differences in spectral information. In this research,Fourier transform near-infrared (FT-NIR) spectral information aboutimparity species of porcini mushroom stipe and cap was collected andcombined into four data matrices. FT-NIR spectra of four data setswere combined with chemometric methods and machine learning for accurateevaluation and identification of different porcini mushroom species.From the results: (1) improved visualization level of t-distributedstochastic neighbor embedding (t-SNE) results after the second derivativepreprocessing compared with raw spectra; (2) after using multiplepretreatment combinations to process the four data matrices, the modelaccuracies based on support vector machine and partial least-squarediscriminant analysis (PLS-DA) under the best preprocessing methodwere 98.73-99.04% and 98.73-99.68%, respectively; (3)by comparing the modeling results of FT-NIR spectra with differentdata matrices, it was found that the PLS-DA model based on low-leveldata fusion has the highest accuracy (99.68%), but residual neuralnetwork (ResNet) model based on the stipe, cap, and average spectraldata matrix worked better (100% accuracy). The above results suggestthat distinct models should be selected for dissimilar spectral datamatrices of porcini mushrooms. Additionally, FT-NIR spectra have theadvantages of being nondevastate and fast; this method is expectedto be a promising analytical tool in food safety control.

  • 相关文献
作者其他论文 更多>>
置顶
购物车
反馈

© 京ICP备09089781号-29 主办单位：云南省农业科学院

学术资源： 中国农业科学院 农业专业知识服务系统 农科联盟资源共建共享平台 中国农业科技文献与信息服务平台 中国农业期刊集成服务平台