食品与发酵工业 >

2026 , Vol. 52 >Issue 11: 358 - 367

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.044343

分析与检测

基于近红外光谱分析的西梅贮藏期定性判别和品质预测

  • 王允海 ,
  • 马文强 ,
  • 项斌斌 ,
  • 崔宽波 ,
  • 杨莉玲 ,
  • 马文杰 ,
  • 何美玲 ,
  • 祝兆帅 ,
  • 孙俪娜
展开
  • 1(新疆大学 机械工程学院, 新疆 乌鲁木齐, 830049)
    2(新疆维吾尔自治区农业科学院农业装备研究所, 新疆 乌鲁木齐, 830091)
第一作者:硕士研究生(马文强研究员和项斌斌高级工程师为共同通信作者,E-mail:mwq4530@163.com;xiangbinbin031@163.com)

收稿日期: 2025-08-27

  修回日期: 2025-11-13

  网络出版日期: 2026-06-24

基金资助

新疆维吾尔自治区农业科学院农业科技创新稳定支持项目(xjnkywdzc-2025003-08-3);新疆维吾尔自治区重点研发项目(2022B02018-4);国家自然科学基金项目(32460613)

收起

Qualitative discrimination and quality prediction model of plum storage period based on near infrared

  • WANG Yunhai ,
  • MA Wenqiang ,
  • XIANG Binbin ,
  • CUI Kuanbo ,
  • YANG Liling ,
  • MA Wenjie ,
  • HE Meiling ,
  • ZHU Zhaoshuai ,
  • SUN Lina
Expand
  • 1(College of Mechanical Engineering, Xinjiang University, Urumqi 830049, China)
    2(Institute of Agricultural Equipment, Xinjiang Academy of Agricultural Sciences, Urumqi 830091, China)

Received date: 2025-08-27

  Revised date: 2025-11-13

  Online published: 2026-06-24

Fold

摘要

为实现贮藏阶段关键品质指标的定性判别及定量预测,该研究以贮藏期‘法兰西’西梅为对象,通过采集西梅在(0、15、30、45、60 d)贮藏期的近红外光谱,结合归一化、卷积平滑(Savitzky-Golay smoothing,SG)、多元散射校正(multiplicative scatter correction,MSC)和标准正态变换(standard normal variate,SNV)等预处理方法,及竞争性自适应重加权采样(competitive adaptive reweighted sampling,CARS)、无信息变量消除算法(uninformative variables elimination,UVE)、遗传算法(genetic algorithm,GA)等特征提取方法,分别构建支持向量机、K最近邻、线性判别分析(linear discriminant analysis,LDA)的贮藏期定性判别模型,与偏最小二乘回归(partial least squares regression,PLSR)的可溶性固形物含量、含水率、硬度单品质基准预测模型。在此基础上,融合可溶性固形物含量、含水率和硬度的表征波长,引入了可解释性机器学习SHAP(SHapley Additive exPlanations)方法,并将其应用于融合后特征波长的贡献度评估和分阶段筛选,构建优化的硬度融合预测模型。结果表明,构建的MSC-GA-LDA贮藏阶段定性判别模型在测试集上100%准确判别(测试集样本量n=75,涵盖全部贮藏阶段);构建的单品质基准预测模型中,可溶性固形物含量、含水率、硬度模型分别筛选出30、54、15个关键响应特征,对应模型的最优表现分别为SNV-CARS-PLSR(R2p=0.922 2,RMSEp=0.663 9%)、SNV-GA-PLSR(R2p=0.809 2,RMSEp=1.806e-02%)、SNV-CARS-PLSR(R2p=0.809 0,RMSEp=0.103 6 N);经SHAP筛选的融合模型显著提升硬度预测效果(R2p=0.848 5,RMSEp=0.097 5 N)。研究结果为西梅贮藏品质的无损检测提供了理论依据与技术支持。

关键词： ‘法兰西’西梅; 近红外光谱; 贮藏期; 特征波长; 融合模型

本文引用格式

王允海 , 马文强 , 项斌斌 , 崔宽波 , 杨莉玲 , 马文杰 , 何美玲 , 祝兆帅 , 孙俪娜 . 基于近红外光谱分析的西梅贮藏期定性判别和品质预测[J]. 食品与发酵工业, 2026 , 52(11) : 358 -367 . DOI: 10.13995/j.cnki.11-1802/ts.044343

Abstract

In order to realize the qualitative discrimination and quantitative prediction of key quality indexes in the storage stage, the ‘France’ plum during the storage period was taken as the object in this study.The near-infrared spectra of plum during the storage period of 0, 15, 30, 45 and 60 days were collected.Combined with pretreatment methods such as normalization, convolution smoothing, multiple scattering correction and standard normal transformation, and feature extraction methods such as competitive adaptive reweighted sampling, uninformative variable elimination and genetic algorithm, the qualitative discrimination models of support vector machine, K nearest neighbor and linear discriminant analysis were constructed respectively.Single quality benchmark prediction model of soluble solids content, moisture content and hardness with partial least squares regression.On this basis, the interpretable machine learning SHapley Additive exPlanations(SHAP) method is introduced by combining the characterization wavelengths of soluble solids content and water content with the characterization wavelengths of fruit firmness, and it was applied to the contribution evaluation and phased screening of the fused characteristic wavelengths, and an optimized fruit firmness fusion prediction model is constructed.The results showed that the qualitative discriminant model of MSC-GA-LDA storage stage was 100% accurate on the test set (the sample size of the test set n=75, covering all storage stages).In the single quality benchmark prediction model, 30,54 and 15 key response characteristics were screened out from the soluble solids content, water content and fruit firmness models, respectively.The optimal performances of the corresponding models were SNV-CARS-PLSR (R2p=0.922 2, RMSEp=0.663 9%), SNV-GA-PLSR (R2p=0.809 2, RMSEp=1.806 e-02%) and SNV-CARS-PLSR (R2p=0.809 0, RMSEp=0.103 6 N).The fusion model screened by SHAP significantly improved the prediction effect of fruit hardness (R2p=0.848 5, RMSEp=0.097 5 N), which provided theoretical basis and technical support for the non-destructive detection of storage quality of plums.

Key words： ‘France’ plum; near infrared spectroscopy; storage period; characteristic wavelength; fusion model

参考文献

[1] 晏文丽, 代宇琪,韩淼, 等.新疆西梅采后生理变化与贮运保鲜技术研究进展[J].食品工业,2024,45(4):173-177.
YAN W L, DAI Y Q, HAN M, et al.A review on postharvest physiological and biochemical changes, storage and preservation of Xinjiang Prunus domestica L.Ximei[J].The Food Industry, 2024, 45(4):173-177.
[2] 毕凯悦,阿依则巴·艾尼玩尔,海子悦,等.1-MCP处理对西梅果实贮藏期品质及活性氧代谢的影响[J].食品工业科技,2025,46(14):395-403.
BI K Y, AIZHEBA A, HAI Z Y, et al.Effect of 1-MCP treatment on quality and reactive oxygen metabolism of prune fruits during storage period[J].Science and Technology of Food Industry, 2025,46(14):395-403.
[3] 毛欣然,夏静静,徐惟馨,等.手持式近红外光谱仪测定梨三种品质指标通用模型建模方法研究[J].光谱学与光谱分析,2024,44(2):406-412.
MAO X R, XIA J J, XU W X, et al.Study on modeling method of general model for measuring three quality indexes of pear by handheld near-infrared spectrometer[J].Spectroscopy and Spectral Analysis, 2024,44(2):406-412.
[4] 肖徐,袁进,李静,等.基于异常值双重剔除与改进PLS算法的近红外光谱红缨子糯高粱关键指标快速检测模型优化研究[J].食品与发酵工业, 2025,51(24):368-374.
XIAO X, YUAN J, LI J, et al.Optimization of a rapid test model for key indicators of glutinous sorghum cultivar “Hongyingzi” in near-infrared spectroscopy based on outlier double rejection and improved PLS algorithm[J].Food and Fermentation Industries, 2025,51(24):368-374.
[5] GUO Z M, WANG M M, SHUJAT A, et al.Nondestructive monitoring storage quality of apples at different temperatures by near-infrared transmittance spectroscopy[J].Food Science & Nutrition, 2020, 8(7):3793-3805.
[6] 罗东杰,王勐,张小栓,等.基于Vis/NIR光谱传感的鲜食葡萄糖度检测系统[J].光谱学与光谱分析,2023,43(7):2146-2152.
LUO D J, WANG M, ZHANG X S, et al.Vis/NIR based spectral sensing for SSC of table grapes[J].Spectroscopy and Spectral Analysis, 2023,43(7):2146-2152.
[7] BLAKEY R J, BOWER J P, BERTLING I.Influence of water and ABA supply on the ripening pattern of avocado (Persea americana Mill.) fruit and the prediction of water content using Near Infrared Spectroscopy[J].Postharvest Biology and Technology, 2009, 53(1-2):72-76.
[8] LI X D, DUNKIN F, DEZERT J.Multi-source information fusion:Progress and future[J].Chinese Journal of Aeronautics, 2024, 37(7):24-58.
[9] 高升,王巧华.基于可见/近红外透射光谱技术的红提糖度和含水率无损检测[J].中国光学,2021,14(3):566-577.
GAO S, WANG Q H.Non-destructive testing of red globe grape sugar content and moisture content based on visible/near infrared spectroscopy transmission technology[J].Chinese Optics,2021,14(3):566-577.
[10] 王鑫磊,韩鲁佳,杨增玲,等.预处理方法对异源近红外秸秆营养成分速测模型的影响[J].农业工程学报, 2025,41(9):216-226.
WANG X L, HAN L J, YANG Z L, et al.Influence of preprocessing methods on the rapid quantitative models for crop straw nutrients using different near-infrared spectroscopy devices data[J].Transactions of the Chinese Society of Agricultural Engineering, 2025,41(9):216-226.
[11] 李进发,向阳,王宇,等.基于近红外光谱技术建立刺糖中多糖含量的快速检测模型[J].食品与发酵工业, 2025,51(24):342-348.
LI J F, XIANG Y, WANG Y, et al.Establishment of a rapid detection model for polysaccharide content in Alhagi sparsifolia honey based on near-infrared spectroscopy technology[J].Food and Fermentation Industries, 2025,51(24):342-348.
[12] 谢玉玉,陈志慧,侯雪玲,等.近红外光谱结合化学计量学快速测定阿里红饮片中齿孔酸含量[J].光谱学与光谱分析,2024,44(10):2981-2987.
XIE Y Y, CHEN Z H, HOU X L, et al.A new method for determination of eburicoic acid in Fomes officinalis Ames by NIR combined with PLS[J].Spectroscopy and Spectral Analysis, 2024,44(10):2981-2987.
[13] 于欣冉,赵鹏,宦克为,等.基于GA-SVR的近红外无创检测智能算法研究[J].光谱学与光谱分析,2024,44(11):3020-3028.
YU X R, ZHAO P, HUAN K W, et al.Research on intelligent algorithm of near-infrared spectroscopy non-invasive detection based on GA-SVR method[J].Spectroscopy and Spectral Analysis, 2024,44(11):3020-3028.
[14] HANCOCK J T, KHOSHGOFTAAR T M, LIANG Q X.A problem-agnostic approach to feature selection and analysis using SHAP[J].Journal of Big Data, 2025, 12(1):12.
[15] 石吉勇,李文亭,邹小波,等.基于近红外光谱特征的三文鱼品质多指标快速检测[J].光谱学与光谱分析,2019,39(7):2244-2249.
SHI J Y, LI W T, ZOU, X B, et al.Multi-index rapid detection of salmon quality based on near-infrared spectroscopy[J].Spectroscopy and Spectral Analysis, 2019,39(7):2244-2249.
[16] 张淑芳,雷蕾,雷顺新,等.近红外漫反射光谱的茉莉花产地溯源[J].光谱学与光谱分析,2023,43(11):3389-3395.
ZHANG S F, LEI L, LEI S X, et al.Traceability of geographical origin of jasmine based on near infrared diffuse reflectance spectroscopy[J].Spectroscopy and Spectral Analysis, 2023,43(11):3389-3395.
[17] 李晋,张琛,刘红,等.近红外光谱联合化学计量学在柑橘类水果质量无损检测方面的最新研究及应用进展[J].食品与发酵工业,2024,50(5):367-379.
LI J, ZHANG C, LIU H, et al.Advances on quality detection of citrus fruits by near-infrared spectroscopy combined with chemometrics[J].Food and Fermentation Industries, 2024,50(5):367-379.
[18] FATCHURRAHMAN D, NOSRATI M, AMODIO M L, et al.Comparison performance of visible-NIR and near-infrared hyperspectral imaging for prediction of nutritional quality of goji berry (Lycium barbarum L.)[J].Foods, 2021, 10(7):1676.
[19] 王冬,冯海智,李龙,等.两种近红外光谱仪的番茄可溶性固形物含量定量模型比较研究[J].光谱学与光谱分析,2023,43(5):1351-1357.
WANG D, FENG H Z, LI L, et al.Compare of the quantitative models of SSC in tomato by two types of NIR spectrometers[J].Spectroscopy and Spectral Analysis, 2023,43(5):1351-1357.
[20] 盛晓慧,李子文,李宗朋,等.基于近红外光谱分析技术测定库尔勒香梨硬度[J].光谱学与光谱分析,2019,39(9):2818-2822.
SHENG X H, LI Z W, LI Z P, et al.Determination of Korla pear hardness based on near-infrared spectroscopy[J].Spectroscopy and Spectral Analysis, 2019,39(9):2818-2822.
[21] 高明,杨瑞芳,赵南京,等.基于CNN的地下水多组分多环芳烃3DEEM快速定性定量分析方法[J].光学学报,2025,45(6):387-398.
GAO M, YANG R F, ZHAO N J, et al.Rapid qualitative and quantitative analysis method of multi-component polycyclic aromatic hydrocarbons in groundwater using CNN and 3DEEM[J].Acta Optica Sinica,2025,45(6):387-398.
[22] 许诗咏,石婷,赵玉霞,等.红嘴蓝喜鹊优化器-BP神经网络结合紫外光谱定性识别青稞酒[J].食品与发酵工业, 2025,51(21):346-352.
XU S Y, SHI T, ZHAO Y X, et al.Identification of Qingke liquor qualitatively with the combination of red-billed blue magpie optimizer-BP neural network and ultraviolet spectrum[J].Food and Fermentation Industries, 2025,51(21):346-352.
[23] TANGORRA F M, LOPEZ A, IGHINA E, et al.Handheld NIR spectroscopy combined with a hybrid LDA-SVM model for fast classification of retail milk[J].Foods, 2024, 13(22):3577.
[24] 马俊杰,白梦涵,哦哈尔·帕孜力江,等.臭氧处理对西梅果实采后贮藏AsA-GSH循环及细胞膜脂过氧化的影响[J].食品工业科技, 2026,47(8):368-376.
MA J J, BAI M H, OHAER P, et al.Effects of ozone treatment on AsA-GSH cycle and membrane lipid peroxidation in postharvest storage of prunus fruits[J].Science and Technology of Food Industry,2026,47(8):368-376.
[25] 文钰,祝兆帅,白梦涵,等.近冰温结合气调贮藏对西梅果实货架期品质的影响[J].食品科技,2025,50(5):53-59.
WEN Y, ZHU Z S, BAI M H, et al.Effect of near freezing temperature combined with air-conditioned storage on the shelf-life quality of prunes[J].Food Science and Technology, 2025,50(5):53-59.
[26] 张路思,孙祥杰,宋尚健,等.不同浓度CO2气调贮藏条件下西梅果实品质和生理的变化[J].现代食品科技, 2025, 41(8):140-147.
ZHANG L S, SUN X J, SONG S J, et al.Changes in the fruit quality and physiology of plum fruit under controlled atmosphere storage with different CO2 concentrations[J].Modern Food Science and Technology, 2025, 41(8):140-147.
[27] 曾明飞,朱玉杰,冯国红,等.基于可见/近红外光谱的蓝莓新鲜度快速评价[J].食品与发酵工业,2022,48(20):252-259.
ZENG M F, ZHU Y J, FENG G H, et al.Rapid evaluation of blueberry freshness based on visible/near-infrared spectroscopy[J].Food and Fermentation Industries, 2022,48(20):252-259.
[28] CHEN C, QI S M, ZHANG S S, et al.Study on the quality change and regulation mechanism of ‘shannongsu’ pear under low-temperature storage[J].International Journal of Molecular Sciences, 2025, 26(7):2900.
[29] 王建旭,谭银雨,覃丹,等.基于近红外反射光谱的宣纸含水率无损检测研究[J].光谱学与光谱分析,2025,45(6):1629-1638.
WANG J X, TAN Y Y, QIN D, et al.Research on non-destructive detection of moisture content in Xuan paper based on near-infrared reflectance spectroscopy[J].Spectroscopy and Spectral Analysis, 2025,45(6):1629-1638.
[30] 赖建平,赵辉,王东升,等.GA-XGBoost模型对路基压实质量的预测[J].哈尔滨工业大学学报,2025,57(7):33-41.
LAI J P, ZHAO H, WANG D S, et al.Prediction of dynamic deformation modulus of subgrade based on GA-XGBoost model[J].Journal of Harbin Institute of Technology, 2025,57(7):33-41.
[31] 刘燕德,张雨,姜小刚,等.不同贮藏期水蜜桃硬度及糖度的检测研究[J].光谱学与光谱分析,2021,41(1):243-249.
LIU Y D, ZHANG Y, JIANG X G, et al.Detection on firmness and soluble solid content of peach during different storage days[J].Spectroscopy and Spectral Analysis, 2021,41(1):243-249.
[32] BAPTISTA M L, GOEBEL K, HENRIQUES E M P.Relation between prognostics predictor evaluation metrics and local interpretability SHAP values[J].Artificial Intelligence, 2022, 306:103667.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司