Effects of Scalable One-pass Self-representation Learning on Near Infrared Spectroscopy Regression Modeling

GUO Tuo; LIANG Xiao-juan; MA Jin-fang; YUAN Kai; GE Fa-huan; XIAO Huan-xian

doi:10.19969/j.fxcsxb.22040203

您当前的位置：

首页 >

文章列表页 >

Effects of Scalable One-pass Self-representation Learning on Near Infrared Spectroscopy Regression Modeling

Experimental Techniques and Methods | 更新时间：2023-01-04

- Effects of Scalable One-pass Self-representation Learning on Near Infrared Spectroscopy Regression Modeling
- Journal of Instrumental Analysis Vol. 41, Issue 8, Pages: 1214-1220(2022)
- 作者机构：
  
  1.陕西科技大学电子信息与人工智能学院，陕西西安 710021
  2.暨南大学光电工程系，广东广州 510632
  3.中山大学南沙研究院，广东广州 511458
  4.江西保利制药有限公司，江西赣州 341900
- 作者简介：
- 基金信息：
- DOI：10.19969/j.fxcsxb.22040203
  CLC：
扫描看全文
郭拓,梁小娟,马晋芳等.基于可扩展的自表示学习波段选择算法在近红外光谱回归建模中的影响研究[J].分析测试学报,2022,41(08):1214-1220.

GUO Tuo,LIANG Xiao-juan,MA Jin-fang,et al.Effects of Scalable One-pass Self-representation Learning on Near Infrared Spectroscopy Regression Modeling[J].Journal of Instrumental Analysis,2022,41(08):1214-1220.
郭拓,梁小娟,马晋芳等.基于可扩展的自表示学习波段选择算法在近红外光谱回归建模中的影响研究[J].分析测试学报,2022,41(08):1214-1220. DOI： 10.19969/j.fxcsxb.22040203.

GUO Tuo,LIANG Xiao-juan,MA Jin-fang,et al.Effects of Scalable One-pass Self-representation Learning on Near Infrared Spectroscopy Regression Modeling[J].Journal of Instrumental Analysis,2022,41(08):1214-1220. DOI： 10.19969/j.fxcsxb.22040203.

摘要

该文提出了一种基于可扩展的自表示学习（SOP－SRL）波段选择与偏最小二乘（PLS）建模的定量模型分析方法，以安胎丸指标含量阿魏酸、黄芩苷和汉黄芩苷为研究对象，通过SOP－SRL选取代表性波段，采用PLS建立近红外光谱回归模型，并与相关系数法（CC）、正则化自表示学习算法（RSR）和稀疏子空间聚类法（SSC）3种波段选择算法的建模结果进行对比，以校正决定系数（, $R_{c}^{2}$ ,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37288985&type=,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37288982&type=,2.96333337,3.47133350,）、校正均方根误差（RMSECV）、预测决定系数（, $R_{p}^{2}$ ,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37288992&type=,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37288988&type=,2.96333337,3.47133350,）和预测均方根误差（RMSEP）为评价标准，对回归模型的预测性能进行评估。结果显示，SOP－SRL在3种数据集上均取得了较好的结果，建模波段从全波长的800分别减少到70、67、87；RMSEP分别从0.080 1、6.349 5、0.742 5下降到0.065 3、3.620 8、0.407 3，分别下降了18%、43%、45%；相应的, $R_{p}^{2}$ ,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289000&type=,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37288996&type=,2.96333337,3.47133350,分别从0.911 9、0.879 4、0.915 8提高到0.938 8、0.952 6、0.970 1，分别提高了3%、8%、6%。结果表明，经SOP－SRL波长选择后模型的预测能力相比于其他几种算法得到显著提升，基于SOP－SRL的PLS模型可以实现安胎丸指标含量的快速检测。

Abstract

Near-infrared spectroscopy is widely applied in the quality monitoring process of traditional Chinese medicine since it features with rapid detection，and making no damage to the samples and no pollution to the environment in the meantime. In order to realize the rapid prediction of the target ingredients of Antai Pills，a new near-infrared spectroscopy modeling method was proposed，which combines scalable one-pass self-representation learning（SOP－SRL） with partial least-squares（PLS）. Taking ferulicacid，baicalin and wogonoside in Antai pills as the research objects，the representative bands selected by SOP－SRL were compared with three band selection algorithms，such as correlation coefficient method（CC），regularized self-representation algorithm（RSR） and sparse subspace clustering （SSC）. Then，the quantitative model was established by PLS. The evaluation criteria of the model are root mean squares error of cross validation（RMSECV），corrected determination coefficient（, $R_{c}^{2}$ ,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289008&type=,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289004&type=,3.47133350,4.06400013,），predicted root mean square error（RMSEP） and predicted determination coefficient（, $R_{p}^{2}$ ,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289016&type=,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289012&type=,3.47133350,4.06400013,）. Results indicated that the SOP－SRL had good results on all three datasets. Compared with all bands，the selected bands of SOP－SRL were reduced from 800（FULL） to 70，67 and 87， respectively. The RMSEP decreased from 0.080 1，6.349 5，0.742 5 to 0.065 3，3.620 8，0.407 3，decreased by 18%，43% and 45%，respectively. The , $R_{p}^{2}$ ,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289023&type=,http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=37289020&type=,4.23333359,4.14866638,increased from 0.911 9，0.879 4，0.915 8 to 0.938 8，0.952 6，0.970 1，respectively，increased by 3%，8% and 6%. Therefore，the results of the SOP－SRL algorithm were significantly better than other comparison algorithms. The SOP－SRL algorithm could improve the accuracy of quantative model. The model combining SOP－SRL with PLS could rapidly detect the target ingredients of Antai pills.

关键词

近红外光谱波段选择可扩展的自表示学习方法（SOP－SRL）偏最小二乘法（PLS）指标含量测定

Keywords

near infrared spectroscopywavelength selectionscalable one-pass self-representation learning（SOP－SRL）partial least-squares（PLS）target ingredients determination

references

Yan Y L，Chen B，Zhu D Z．Principles，Technology and Application of NIR Spectral Analysis. Beijing：China Light Industry Press（严衍禄，陈斌，朱大洲．近红外光谱分析的原理、技术与应用. 北京：中国轻工业出版社），2013：24－28.

Chu X L．Practical Handbook of Near Infrared Spectroscopy. Beijing：China Machine Press（褚小立．近红外光谱分析技术实用手册. 北京：机械工业出版社），2016：116－117.

Yang Y，Yang L C，Ji X L，Li Y H，Xu Q R，Tong H B．J. Instrum. Anal. （杨越，杨留长，纪晓亮，李易航，徐倩茹，佟海滨.分析测试学报），2020，39（11）：1311－1319.

Zhao Z，Liu H．Machine Learning，Twenty-fourth International Conference，Corvallis，Oregon，USA，June 20－24，2007.

Zhu G K，Huang Y C，Lei J S，Bi Z Q，Xu F F．IEEE Trans. Geosci. Remote Sens.，2015，54（1）：227－239.

Ahmad M，Haq I U，Mushtaq Q，Sohaib M．Int. J. Eng. Technol.，2011，3（6）：606－614.

Thiagarajan J J，Ramamurthy K N，Spanias A．IEEE Trans. Image Process.，2014，23（7）：2905－2915.

Ma Y C，Zhang Y，Zhang N，Zhu H D．Basic Sci. J. Text. Univ. （马盈仓，张要，张宁，朱恒东．纺织高校基础科学学报），2021，34（3）：102－111，120.

Jian C R，Weng Q．J. Fuzhou Univ.：Nat. Sci. Ed. （简彩仁，翁谦．福州大学学报：自然科学版），2022，50（1）：1－8.

Zhu P F，Zuo W M，Zhang L，Hu Q H，Shiu S C K．Pattern Recognit.，2015，48（2）：438－446.

Wei X H，Zhu W，Liao B，Cai L J．IEEE Trans. Geosci. Remote Sens.，2019，57（7）：4360－4374.

Gao S W，Wang Z M，Hong Z R．Process Autom. Instrum. （高世伟，王忠民，洪梓榕．自动化仪表），2018，39（9）：52－55.

Ma J F，Wang X L，Xiao X，Peng Y，Ge F H．World Sci. Technol.－Mod. Tradit. Chin. Med. （马晋芳，王雪利，肖雪，彭银，葛发欢．世界科学技术－中医药现代化），2018，20（5）：651－659.

Liu C L，Hu Y J，Wu S N，Sun X R，Dou S L，Miao Y Q，Dou Y．J. Food Sci. Technol. 刘翠玲，胡玉君，吴胜男，孙晓荣，窦森磊，苗雨晴，窦颖．食品科学技术学报），2014，32（5）：74－79.

Ma L W，Guo T，Ma J F，Shi Q L，Xiao H X．J. Instrum. Anal. （马力文，郭拓，马晋芳，史庆龙，肖环贤．分析测试学报），2020，39（10）：1254－1259.

Chen S B，Hu Z．Food Ind. （陈素彬，胡振．食品工业），2019，40（12）：329－333.

Lu H X，Zhang J，Li L Q，Liu Z B，Yang H H，Feng Y C，Yin L H．Spectrosc. Spectral Anal. 路皓翔，张静，李灵巧，刘振丙，杨辉华，冯艳春，尹利辉．光谱学与光谱分析），2021，41（6）：1782－1788.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Determination of Polysaccharide Contents during Polygoni Multiflori Concoction by Near Infrared Spectroscopy

Research on Application of Near Infrared Spectroscopy in Situ Detection of Sichuan Ophiopogon Japonicus

Origin Discrimination Model of Crataegi Folium Based on Sparse Principal Component Analysis for Feature Selection

Study on a New Method for Determination of Luteolin Concentration through Water Spectra

Related Author

No data

Related Institution

School of Electronic Information and Artificial Intelligence，Shaanxi University of Science and Technology

HongZhengDao（China） Traditional Chinese Medicine Research Company Ltd

Guangdong Metabolic Diseases Research Center of Integrated Chinese and Western Medicine（Institute of Traditional Chinese Medicine），Guangdong Pharmaceutical University

NMPA Key Laboratory for Rapid Testing Technology of Drugs，Guangdong Institute for Drug Control

National Sugar Engineering and Technology Research Center，Shandong University

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