Quantitative analysis of sorghum starch using machine learning and near-infrared spectroscopy

Authors

• LI Ke, College of Food Science and Technology , Hunan Agricultural University , Changsha , Hunan 410128 , China
• OUYANG Runkun, College of Food Science and Technology , Hunan Agricultural University , Changsha , Hunan 410128 , China
• DAI Yijie, College of Biology and Environmental Engineering , Guiyang University , Guiyang , Guizhou 550005 , China
• LI Jie, College of Food Science and Technology , Hunan Agricultural University , Changsha , Hunan 410128 , China
• HOU Aixiang, College of Food Science and Technology , Hunan Agricultural University , Changsha , Hunan 410128 , China
• YANG Yu, College of Food Science and Technology , Hunan Agricultural University , Changsha , Hunan 410128 , China
• LI Zongjun, College of Food Science and Technology , Hunan Agricultural University , Changsha , Hunan 410128 , ChinaFollow

Corresponding Author(s)

李宗军（1967—），男，湖南农业大学教授，博士。E-mail：hnlizongjun@163.com

Abstract

[Objective] To establish a model for the rapid quantification of sorghum starch based on near-infrared spectroscopy and machine learning algorithms. [Methods] Brewing sorghum was used as the raw material. Near-infrared diffuse reflectance spectra were collected using a near-infrared spectrometer. The Isolation Forest algorithm was employed to remove outlier samples and optimize the dataset. Subsequently, multiple preprocessing methods and regression models were compared using various modeling software to analyze and establish a quantitative model for sorghum starch. [Results] After preprocessing with the Isolation Forest algorithm, the partial least squares regression (PLSR) model exhibited the best performance, with a coefficient of determination for the calibration set (R2c) of 0.993 and a root mean square error of cross-validation (RMSECV) of 3.401. [Conclusion] The model is simple, rapid, and accurate, and is suitable for the quantitative analysis of starch content in sorghum.

Publication Date

7-13-2026

First Page

60

Last Page

66

DOI

10.13652/j.spjx.1003.5788.2025.80865

Recommended Citation

Ke, LI; Runkun, OUYANG; Yijie, DAI; Jie, LI; Aixiang, HOU; Yu, YANG; and Zongjun, LI (2026) "Quantitative analysis of sorghum starch using machine learning and near-infrared spectroscopy," Food and Machinery: Vol. 42: Iss. 6, Article 8.
DOI: 10.13652/j.spjx.1003.5788.2025.80865
Available at: https://www.ifoodmm.cn/journal/vol42/iss6/8

References

[1] 赵珊珊,郭志强,朱立勋,等.高粱高亲和硝酸盐转运蛋白NRT 2/3基因家族鉴定、表达与 DNA变异分析 [J].生物工程学报,2023,39(7):2 743-2 761.ZHAO S S,GUO Z Q,ZHU L H,et al.Identification,expression and DNA variation analysis of high affinity nitrate transporter NRT 2/3 gene family in Sorghum bicolor [J].Chinese Journal of Biotechnology,2023,39(7):2 743-2 761.
[2] XIN Z G,WANG M L,CUEVAS H E,et al.Sorghum genetic,genomic,and breeding resources [J].Planta,2021,254(6):114.
[3] HASHEMI S S,ABBASI-RIYAKHUNI M,DENAYER J F M,et al.Efficient bioremediation of distillery and dairy wastewaters:a three-stage biorefinery for high-quality aquaculture feed and bioenergy generation [J].Process Safety and Environmental Protection,2023,180:566-574.
[4] 向娜娜,王晓琼,孙美珍,等.基于近红外光谱技术快速测定高粱中淀粉及单宁含量的研究 [J].中国饲料,2024 (23):345-349,366.XIANG N N,WANG X Q,SUN M Z,et al.Research on rapid determination of starch and tannin content in sorghum based on near-infrared spectroscopy technique [J].China Feed,2024 (23):345-349,366.
[5] BIAN M H,FANG Y L,YANG K L,et al.Comparative analysis of microbial communities and flavor compounds in fermented grains from different sorghum varieties used in Sichuan Xiaoqu liquor [J].LWT-Food Science and Technology,2025,222:117640.
[6] 陆婉珍.现代近红外光谱分析技术 [M].2版.北京:中国石化出版社,2007:1-11.LU W Z.Modern near infrared spectroscopy analytical technology[M].2nd ed.Beijing:China Petrochemical Press,2007:1-11.
[7] 乔宁,饶敏,黄雪媛,等.基于便携式近红外光谱仪和随机森林方法快速鉴别蜂蜜品种 [J].中国口岸科学技术,2024,6(8):75-80.QIAO N,RAO M,HUANG X Y,et al.Rapid identification of honey varieties based on portable near-infrared spectroscopy and random forest algorithm [J].China Port Science and Technology,2024,6(8):75-80.
[8] 王雪,张广月,马铁民,等.基于近红外光谱的灌浆期玉米籽粒水分定量分析通用模型 [J].农业工程学报,2025,41(8):291-300.WANG X,ZHANG G Y,MA T M,et al.Generalized model for the quantitative moisture analysis of maize grains during filling stage based on near-infrared spectroscopy [J].Transactions of the Chinese Society of Agricultural Engineering,2025,41(8):291-300.
[9] 张云霞,余佶,李运通,等.基于近红外光谱法的馥郁香型白酒基酒中 4种主要有机酸检测模型构建 [J].食品与机械,2025,41(4):72-80.ZHANG Y X,YU J,LI Y T,et al.Detection modelling of four major organic acids in the base wine of Fuyuxiangxing crude Baijiu based on near-infrared spectroscopy [J].Food & Machinery,2025,41(4):72-80.
[10] 李楠,杨春杰.基于近红外光谱技术的小米产地溯源研究[J].食品与机械,2020,36(9):97-101.LI N,YANG C J.Geographic origin determination of millet based on near infrared spectroscopy technique [J].Food & Machinery,2020,36(9):97-101.
[11] NING J M,SHENG M G,YI X Y,et al.Rapid evaluation of soil fertility in tea plantation based on near-infrared spectroscopy [J].Spectroscopy Letters,2018,51(9):463-471.
[12] 姚力,李宗军,朱门君,等.近红外漫反射光谱法快速检测高纤维素、木质素物料水分含量 [J].食品与机械,2024,40(2):69-73.YAO L,LI Z J,ZHU M J,et al.Rapid determination of moisture content of high cellulose and lignin materials by near-infrared diffuse reflectance spectroscopy [J].Food & Machinery,2024,40(2):69-73.
[13] 赵玉霞,张明锦,王茹,等.反向传播神经网络结合紫外—近红外融合光谱对 “互助”青稞酒的判别研究 [J].光谱学与光谱分析,2025,45(5):1 290-1 299.ZHAO Y X,ZHANG M J,WANG R,et al.Discriminative study on Huzhu Qingke liquor by back propagation neural network combined with ultraviolet-near infrared fusion spectroscopy [J].Spectroscopy and Spectral Analysis,2025,45(5):1 290-1 299.
[14] BROWN T,JACOBS B.An introduction to computer science with emphasis on statistics and the social sciences [J].Computers & Education,1978,2(4):267-270.
[15] 侯江霞,姜金辉,王琛鑫,等.基于机器学习的抑菌活性物质筛选研究进展 [J].食品科学,2025,46(14):366-375.HOU J X,JIANG J H,WANG C X,et al.Research progress in the screening of antimicrobial substances based on machine learning [J].Food Science,2025,46(14):366-375.
[16] 黄林峰,蒋雪松,贾志成,等.基于深度学习的梨树养分含量高光谱监测 [J].光谱学与光谱分析,2024,44(12):3 543-3 552.HUANG L F,JIANG X F,JIA Z C,et al.Deep learning-based monitoring of nutrient content in pear trees [J].Spectroscopy and Spectral Analysis,2024,44(12):3 543-3 552.
[17] 宋少忠,刘园园,周紫阳,等.基于高光谱图像技术的高粱品种识别研究 [J].光谱学与光谱分析,2024,44(5):1 392-1 397.SONG S Z,LIU Y Y,ZHOU Z Z,et al.Identification of sorghum breed by hyperspectral image technology [J].Spectroscopy and Spectral Analysis,2024,44(5):1 392-1 397.
[18] FEMENIAS A,BAINOTTI M B,GATIUS F,et al.Standardization of near infrared hyperspectral imaging for wheat single kernel sorting according to deoxynivalenol level[J].Food Research International,2021,139:109925.
[19] ZHANG Y L,DAI A J,WU H Y,et al.Development of a benzothianone-based NIR fluorescent probe and its application of Pd 0 detection in bioimaging and smartphone-assisted water samples [J].Journal of Photochemistry and Photobiology A:Chemistry,2025,469:116576.
[20] LEE Y,CAPPELLATO M,DI CAMILLO B.Machine learning-based feature selection to search stable microbial biomarkers:application to inflammatory bowel disease [J].GigaScience,2022,12:giad 083.
[21] 陈闻鹤,程龙生,常志朋,等.基于 Lasso稳健马田系统的相对贫困识别方法 [J].系统工程理论与实践,2022,42(2):527-544.CHEN W H,CHENG L S,CHANG Z P,et al.Identification of relative poverty based on Lasso-based robust Mahalanobis-Taguchi system [J].Systems Engineering-Theory & Practice,2022,42(2):527-544.
[22] XIONG Z M,ZHU D F,LIU D F,et al.Anomaly detection of metallurgical energy data based on iForest-AE [J].Applied Sciences,2022,12(19):9 977.