Source Identification of <italic style="font-style: italic">Radix Glycyrrhizae</italic>（Licorice） Based on the Fusion of Hyperspectral and Texture Feature

PAN Yi-xia; ZHANG Hong-xu; YAN Ji-zhong; ZHANG Hui

doi:10.12452/j.fxcsxb.24042504

您当前的位置：

首页 >

文章列表页 >

Source Identification of Radix Glycyrrhizae（Licorice） Based on the Fusion of Hyperspectral and Texture Feature

Scientific Papers | 更新时间：2024-11-26

- Source Identification of Radix Glycyrrhizae（Licorice） Based on the Fusion of Hyperspectral and Texture Feature
- Journal of Instrumental Analysis Vol. 43, Issue 11, Pages: 1745-1753(2024)
- 作者机构：
  
  浙江工业大学药学院，浙江杭州 310014
- 作者简介：
- 基金信息：
- DOI：10.12452/j.fxcsxb.24042504
  CLC： O657.3;R284
- Published：15 November 2024，
  
  Received：25 April 2024，
  
  Revised：29 May 2024，
- 稿件说明：
移动端阅览
潘忆瑕,张弘旭,颜继忠,张慧.基于高光谱与纹理特征融合的甘草药材基源鉴别研究[J].分析测试学报,2024,43(11):1745-1753.

PAN Yi-xia,ZHANG Hong-xu,YAN Ji-zhong,ZHANG Hui.Source Identification of Radix Glycyrrhizae（Licorice） Based on the Fusion of Hyperspectral and Texture Feature[J].Journal of Instrumental Analysis,2024,43(11):1745-1753.
潘忆瑕,张弘旭,颜继忠,张慧.基于高光谱与纹理特征融合的甘草药材基源鉴别研究[J].分析测试学报,2024,43(11):1745-1753. DOI： 10.12452/j.fxcsxb.24042504.

PAN Yi-xia,ZHANG Hong-xu,YAN Ji-zhong,ZHANG Hui.Source Identification of Radix Glycyrrhizae（Licorice） Based on the Fusion of Hyperspectral and Texture Feature[J].Journal of Instrumental Analysis,2024,43(11):1745-1753. DOI： 10.12452/j.fxcsxb.24042504.

摘要

不同甘草的基源鉴别对保障其临床应用具有重要意义。该研究基于乌拉尔甘草、胀果甘草和光果甘草3种基源共89根甘草药材的光谱特征，进一步利用灰度行程矩阵从4个角度（0°、45°、90°和135°）分别提取11个纹理特征参数（短行程强调、长行程强调、灰度不均匀性、行程不均匀性、行程百分比、低灰度行程强调、高灰度行程强调、短行程低灰度强调、短行程高灰度强调、长行程低灰度强调、长行程高灰度强调）对甘草药材进行鉴别，开发了归一化后求平均值的差异度法（NAD），筛选出对基源鉴别贡献较大的5个纹理特征，并将其与光谱特征进行融合，以提高基源鉴别的准确性。与传统的特征筛选方法SKB法、RFE法和SFM法相比，归一化后求平均值的差异度法展现出更高的适用性，支持向量机（SVM）分类器在测试集上的基源鉴别准确率达到了95.28%，高于未筛选（92.22%）、SKB法筛选（95.00%）、RFE法筛选（95.00%）和SFM法筛选（93.89%）的准确率。该研究提出的数据挖掘方法充分利用高光谱的光谱数据和图像数据，将光谱特征和纹理特征进行有机融合，为甘草药材的基源鉴别提供了一种新型无损的检测方法。

Abstract

The identification of different sources of licorice is of great significance for ensuring its clinical application. Based on the spectral features of 89 licorice medicinal materials from three sources，including

Glycyrrhiza uralensis

Fisch.，

Glycyrrhiza inflata

Bat.，and

Glycyrrhiza glabra

L.，11 texture feature parameters（short run emphasis，long run emphasis，gray-level non-uniformity，run-length non-uniformity，run percentage，low gray-level run emphasis，high gray-level run emphasis，short run low gray-level emphasis，short run high gray-level emphasis，long run low gray-level emphasis，long run high gray-level emphasis） were extracted using the gray-level run-length matrix from four perspectives（0°，45°，90°，and 135°）. Through an innovative feature selection method-normalized average discrepancy method，texture features with significant contributions were selected and fused with spectral features to improve the accuracy of source identification. Compared to traditional feature selection methods such as SelectKBest（SKB），recursive feature elimination（RFE），and SelectFromModel（SFM），the normalized average discrepancy method demonstrated higher applicability in the dataset of this study，with a source identification accuracy

of 95.28% on the test set using support vector machine（SVM） classifier，higher than the unselected（92.22%），SKB-selected（95.00%），RFE-selected（95.00%），and SFM-selected（93.89%） accuracies. The data mining method proposed in this study fully utilizes the hyperspectral data and image data，integrating spectral features and texture features，providing a new non-destructive analytical techniques for the source identification of licorice medicinal materials.

关键词

甘草基源鉴别高光谱成像技术机器学习灰度行程矩阵特征选择

Keywords

licoricesource identificationhyperspectral imaging technologymachine learninggray-level run-length matrixfeature selection

references

The State Pharmacopoeia Commission of the People's Republic of China. Pharmacopoeia of the People’s Republic of China.Volume I. Beijing：China Medical Science and Technology Press（国家药典编委会. 中华人民共和国药典. 2020年版一部. 北京：中国医药科技出版社），2020：88.

Yu Z B，Dong S H，Zhang J W，Li D Y，Liu Y，Li H，Du S S. J. Ethnopharmacol.，2022，298：115613.

Ding Y M，Brand E，Wang W Q，Zhao Z Z. J. Ethnopharmacol.，2022，298：115594.

Sun P P，Yu P，Ge S Y. Chin. J. Rural Med. Pharm.（孙萍萍，余平，葛淑瑜. 中国乡村医药），2023，30（20）：23-25.

Liu J. Inn. Mong. J. Tradit. Chin. Med.（刘杰. 内蒙古中医药），2024，43（3）：146-147.

Han K，Wang M，Zhang L，Wang C Y. Molecules，2018，23（10）：2728.

Li Y Q，Kong D X，Wu H. Rev. Bras. Farmacogn.，2018，28（5）：533-541.

Zhang H，Xu C，Tian Q H，Zhang Y，Zhang G M，Guan Y X，Tong S Q，Yan J Z. J. Ethnopharmacol.，2021，264：113282.

Zhang H，Wang J N，Chen Y，Shen X W，Jiang H J，Gong X C，Yan J Z. J. Sep. Sci.，2019，42（6）：1117-1298.

Bian X H，Lu Z K，van Kollenburg G. Anal. Methods，2020，12（17）：3499-3507.

Zhen Z F. J. Sichuan Tradit. Chin. Med.（甄中锋. 四川中医），2018，36（12）：178-180.

Liu Y C，Li Z，Ma L D，Guo H Z，Chen Y G，Fu X T，Wang W Q. Chin. Pharm. J.（刘育辰，李铮，马丽端，郭洪祝，陈有根，傅欣彤，王文全.中国药学杂志），2011，46（14）：1112-1116.

Li T，Qiao Z X，Li M，Zhou N，Ren G X，Jiang D，Liu C S. Int. J. Food Prop.，2023，26（1）：197-207.

Zhang H，Gao S H. J. Instrum. Anal.（张浩，高树辉. 分析测试学报），2023，42（7）：817-824.

Ma L，Ma Q，Li Y J，Zhang Y Y，Wang J，Ma S Y，Ma Y，Du M H，Wu L G. J. Instrum. Anal.（马玲，马倩，李亚娇，张祎洋，王静，马思艳，马燕，杜明华，吴龙国. 分析测试学报），2023，42（2）：210-215.

Pan Y X，Zhang H X，Chen Y，Gong X C，Yan J Z，Zhang H. Crit. Rev. Anal. Chem.，2023. https：//doi.org/10.1080/10408347.2023.2207652https://doi.org/10.1080/10408347.2023.2207652.

Yin W J，Ru C L，Zheng J，Zhang L，Yan J Z，Zhang H. China J. Chin. Mater. Med.（殷文俊，茹晨雷，郑洁，张璐，颜继忠，张慧. 中国中药杂志），2021，46（4）：923-930.

Zhang H，Pan Y X，Chen Y，Zhang H X，Xie J H，Gong X C，Zhu J Q，Yan J Z. Analyst，2024，149：1837-1848.

Zhao X X，Ruan J J，Li J，Dai C X，Pei M C，Zhou Y. Magn. Reson. Imaging，2024，110：7-16.

Guyon I，Elisseeff A. J. Mach. Learn. Res.，2003，3：1157-1182.

1.4.2 Scikit-Learn（2024）. 1.13.2 Univariate Feature Selection. ［2024-5-20］. https：//scikit-learn.org/stable/modules/feature_selection.html#univariate-feature-selectionhttps://scikit-learn.org/stable/modules/feature_selection.html#univariate-feature-selection.

Guyon I，Weston J，Barnhill S，Vapnik V. Mach. Learn，2002，46：389-422.

1.4.2 Scikit-Learn（2024）. 1.13.3 Recursive Feature Elimination. ［2024-5-20］. https：//scikit-learn.org/stable/modules/feature_selection.html#recursive-feature-eliminationhttps://scikit-learn.org/stable/modules/feature_selection.html#recursive-feature-elimination.

Krupinova J A，Elfimova A R，Rebrova O Y，Voronkova I A，Eremkina A K，Kovaleva E V，Maganeva I S，Gorbacheva A M，Bibik E E，Deviatkin A A，Melnichenko G A，Mokrysheva N G. J. Pathol. Inf.，2022，13：100134.

1.4.2 Scikit-Learn（2024）. 1.13.4 Feature Selection Using SelectFromModel. ［2024-5-20］. https：//scikit-learn.org/stable/modules/feature_selection.html#feature-selection-using-selectfrommodelhttps://scikit-learn.org/stable/modules/feature_selection.html#feature-selection-using-selectfrommodel.

Belhumeur P N，Hespanha J P，Kriegman D J. IEEE Trans. Pattern Anal. Mach. Intell.，1997，19（7）：711-720.

Sun J L，Liu P，Zhai H W. J. Hebei Acad. Sci.（孙洁丽，刘沛，翟浩文. 河北省科学院学报），2022，39（5）：1-6.

Vapnik V N，Lerner A Y. Avtom. Telemekh.，1963，24（6）：774-780.

Ozaki Y. Anal. Sci.，2021，37（9）：1193-1212.

Tomar M，Bhardwaj R，Kumar M，Sumer P S，Krishnan V，Kansal R，Verma R，Yadav V K，Dahuja A，Ahlawat S P，Rana J C，Satyavathi C T，Praveen S，Sachdev A. LWT-Food Sci. Technol.，2021，149：111813.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on Rapid Prediction of Microbial Limit of Intermediates in Traditional Chinese Medicine Formulations Based on Machine Learning

Recent Advances in the Application of Machine Learning in Environmental Analysis and Detection

Related Author

WANG Hai-xia

LI Zheng

LIU Yi-dan

LI Zhi-yao

ZHAO Yu-wen

ZHANG Bo

LIU Si-si

Related Institution

Haihe Laboratory of Modern Chinese Medicine

State Key Laboratory on Technologies for Chinese Medicine Pharmaceutical Process，Control and Intelligent Manufacture，Jiangsu Kanion Pharmaceutical Co.，Ltd

Tianjin Key Laboratory of Intelligent and Green Pharmaceuticals for Traditional Chinese Medicine

College of Pharmaceutical Engineering of Traditional Chinese Medicine，Tianjin University of Traditional Chinese Medicine

State Key Laboratory on Technologies for Chinese Medicine Pharmaceutical Process；Control and Intelligent Manufacture，Jiangsu Kanion Pharmaceutical Co.，Ltd

Address：100 Xianlie Middle Road, Guangzhou Postal code：510070
Tel：020-87684776,37656606 Fax：020-87684776 Email：fxcsxb@china.com
Technical support is provided by Beijing Founder electronics co., LTD 粤ICP备12017312号-3 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰

Source Identification of Radix Glycyrrhizae（Licorice） Based on the Fusion of Hyperspectral and Texture Feature

DOI：10.12452/j.fxcsxb.24042504

摘要

Abstract

关键词

Keywords

references