食品与发酵工业 >

2023 , Vol. 49 >Issue 13: 31 - 39

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.034609

研究报告

基于高光谱成像的非包装和PE包装冷鲜猪大排肉中热杀索丝菌的含量预测

  • 章泽华 ,
  • 刘小花 ,
  • 兰维杰 ,
  • 唐长波 ,
  • 屠康 ,
  • 吴菊清 ,
  • 武杰 ,
  • 潘磊庆
展开
  • 1(南京农业大学 食品科技学院,江苏 南京,210095)
    2(蚌埠学院 食品与生物工程学院,安徽 蚌埠,233030)
第一作者:硕士研究生(潘磊庆教授为通信作者, E-mail:pan_leiqing@njau.edu.cn)

收稿日期: 2022-12-13

  修回日期: 2023-01-28

  网络出版日期: 2023-08-07

基金资助

江苏省重点研发计划项目(BE2020693);国际合作项目(2019YFE0103800);国家重点研发项目(2021YFC2101400);国家自然基金(32272345);安徽省科技重大专项(201903a06020010);安徽省长三角科技创新联合攻关专项(202004g01020009)

收起

Prediction of Brochothrix thermosphacta in unpackaged and PE-packaged chilled pork chops based on hyperspectral imaging

  • ZHANG Zehua ,
  • LIU Xiaohua ,
  • LAN Weijie ,
  • TANG Changbo ,
  • TU Kang ,
  • WU Juqing ,
  • WU Jie ,
  • PAN Leiqing
Expand
  • 1(College of Food Science and Technology, Nanjing Agricultural University, Nanjing 210095, China)
    2(School of Food and Biological Engineering, Bengbu University, Bengbu 233030, China)

Received date: 2022-12-13

  Revised date: 2023-01-28

  Online published: 2023-08-07

Fold

摘要

为探究高光谱成像技术(hyperspectral imaging,HSI)在包装冷鲜肉微生物检测上的适用性,提出了一种基于HSI的包装冷鲜猪大排肉中热杀索丝菌含量的预测方法。采集接种热杀索丝菌的非包装和聚乙烯(polyethylene,PE)包装冷鲜猪大排肉在400~1 000 nm和1 000~2 000 nm波段内的HSI数据,选择不同的预处理算法进行光谱预处理,再通过连续投影算法(successive projections algorithm,SPA)和竞争性自适应重加权算法提取特征波长,分别基于全波段和特征波长建立热杀索丝菌含量预测的偏最小二乘法(partial least squares,PLS)和支持向量机(support vector machine,SVM)模型。结果表明,PE包装组样品的光谱响应值略小于非包装组,但不影响建模效果。基于400~1 000 nm内全波段和特征波段构建的非包装和PE包装冷鲜猪大排肉中热杀索丝菌预测模型优于1 000~2 000 nm内的。其中,基于SPA算法筛选的特征波长建立的预测模型在最大限度减少波段的同时保证了较高的预测精度,非包装组最优模型为1 st-SPA-SVM(RP2=0.932, RPD=3.674),PE包装组最优模型为OSC-SPA-PLS(RP2=0.919, RPD=3.537),这为HSI技术应用于PE包装冷鲜肉中微生物的检测提供了方法参考和数据支撑。

关键词： PE包装冷鲜猪大排肉; 热杀索丝菌; 高光谱成像; 支持向量机; 偏最小二乘法

本文引用格式

章泽华 , 刘小花 , 兰维杰 , 唐长波 , 屠康 , 吴菊清 , 武杰 , 潘磊庆 . 基于高光谱成像的非包装和PE包装冷鲜猪大排肉中热杀索丝菌的含量预测[J]. 食品与发酵工业, 2023 , 49(13) : 31 -39 . DOI: 10.13995/j.cnki.11-1802/ts.034609

Abstract

To investigate the applicability of hyperspectral imaging (HSI) for microbial detection of packaged chilled meat, a method was proposed for quantifying Brochothrix thermosphacta in packaged chilled pork chops by HSI. The HSI data in the 400-1 000 nm and 1 000-2 000 nm were collected from unpackaged and polyethylene (PE) packaged chilled pork chops inoculated with B. thermosphacta, and preprocessed using different preprocessing algorithms, followed by the extraction of feature wavelengths by the successive projection algorithm (SPA) and the competitive adaptive reweighting sampling (CARS). Then, the partial least squares (PLS) and support vector machine (SVM) models for predicting B. thermosphacta contents were developed based on the full-band and the characteristic wavelengths, respectively. The results showed that the spectral values of the PE-packaged samples were slightly lower than those of the unpackaged group, without affecting the accuracy of the model. The prediction models based on full-band and feature wavelengths within 400-1 000 nm for B. thermosphacta in unpackaged and PE-packaged chilled pork chops were superior to those within 1 000-2 000 nm, where prediction models based on the feature wavelengths screened by the SPA algorithm ensured high prediction accuracy while minimizing the wavelengths, and the optimal models were the 1 st-SPA- SVM model (RP2=0.932, RPD=3.674) for the unpackaged group and OSC-SPA-PLS model (RP2=0.919, RPD=3.537) for the PE-packaged group, respectively. It provides the methodological reference and data support for the application of HSI technology for the microbial detection in PE-packaged chilled meat.

Key words： PE-packaged chilled pork chops; Brochothrix thermosphacta; hyperspectral imaging; support vector machine; partial least squares

参考文献

[1] 邓森荣, 向蓉, 魏琦麟, 等.冷鲜肉微生物种类及微生物预测模型的研究进展[J].广东农业科学, 2021, 48(3):133-141.
DENG S R, XIANG R, WEI Q L, et al.Research progress in microbial species and microbial prediction model of chilled meat[J].Guangdong Agricultural Sciences, 2021, 48(3):133-141.
[2] 何鸿举, 蒋圣启, 马汉军, 等.基于NIR高光谱技术快速预测冷鲜鸡肉热杀索丝菌含量[J].食品工业科技, 2020, 41(13):241-246;252.
HE H J, JIANG S Q, MA H J, et al.NIR hyperspectral imaging technology for rapid prediction of Brochothrix thermosphacta in fresh chilled chicken[J].Science and Technology of Food Industry, 2020, 41(13):241-246;252.
[3] 张婧.平板计数法与纸片法检测食品微生物菌落总数的比较分析[J].中国食品, 2021(3):92.
ZHANG J.Comparative analysis of microbial colony count of food by plate counting and paper tablet method[J].China Food, 2021(3):92.
[4] 张建军, 唐轶君, 李晶, 等.基于测试片法与平板计数法对不同类型食品中菌落总数结果的比较分析[J].食品安全质量检测学报, 2021, 12(14):5606-5612.
ZHANG J J, TANG Y J, LI J, et al.Comparative analysis of the results of total number of colonies in different types of food based on test piece method and plate counting method[J].Journal of Food Safety & Quality, 2021, 12(14):5606-5612.
[5] 周慧, 支竹伟, 刘涵, 等.多重实时荧光定量PCR检测肴肉中的特定腐败菌[J].粮食与食品工业, 2014, 21(3):86-89;92.
ZHOU H, ZHI Z W, LIU H, et al.Multiplex real-time quantitive PCR detection of specific spoilage organisms in Yao meat[J].Cereal & Food Industry, 2014, 21(3):86-89;92.
[6] 马骥. 基于高光谱成像技术的加工猪肉品质快速检测方法研究[D].广州:华南理工大学, 2019.
MA J.Research on rapid detection methods for the quality of processed pork based on hyperspectral imaging technology[J].Guangzhou:South China University of Technology, 2019.
[7] 曲佳欢, 马骥, 孙大文, 等.高光谱成像在食品质量评估方面的研究进展与应用(一)[J].肉类研究, 2012, 26(4):40-45.
QU J H, MA J, SUN D W, et al.Researches and applications of hyperspectral imaging in food industry(Ⅰ)[J].Meat Research, 2012, 26(4):40-45.
[8] TAO F F, PENG Y K.A nondestructive method for prediction of total viable count in pork meat by hyperspectral scattering imaging[J].Food and Bioprocess Technology, 2015, 8(1):17-30.
[9] 庄齐斌, 郑晓春, 杨德勇, 等.基于高光谱反射特性的猪肉新鲜度和腐败程度的对比分析[J].食品科学, 2021, 42(16):254-260.
ZHUANG Q B, ZHENG X C, YANG D Y, et al.Comparative analysis of pork freshness and spoilage based on hyperspectral reflection characteristics[J].Food Science, 2021, 42(16):254-260.
[10] ACHATA E M, OLIVEIRA M, ESQUERRE C A, et al.Visible and NIR hyperspectral imaging and chemometrics for prediction of microbial quality of beef Longissimus dorsi muscle under simulated normal and abuse storage conditions[J].LWT, 2020, 128:109463.
[11] 顾欣哲. 基于电子鼻响应信息的猪肉假单胞菌生长预测模型的构建[D].南京:南京农业大学, 2017.
GU X Z.Constructing the predictive growth model of Pseudomonas in pork based on electronic nose[D].Nanjing:Nanjing Agricultural University, 2017.
[12] ZHOU B J, FAN X, SONG J, et al.Growth simulation of Pseudomonas fluorescens in pork using hyperspectral imaging[J].Meat Science, 2022, 188:108767.
[13] LI J L, SUN D W, CHENG J H.Recent advances in nondestructive analytical techniques for determining the total soluble solids in fruits:A review[J].Comprehensive Reviews in Food Science and Food Safety, 2016, 15(5):897-911.
[14] 周孟然, 余道洋, 胡锋, 等.基于连续投影算法的食用油激光诱导荧光光谱特征波长筛选[J].河南师范大学学报(自然科学版), 2021, 49(2):46-53.
ZHOU M R, YU D Y, HU F, et al.Continuous projection algorithm for screening characteristic wavelength of laser induced fluorescence spectrum of edible oil[J].Journal of Henan Normal University (Natural Science Edition), 2021, 49(2):46-53.
[15] 郭城, 吴卫成, 谌迪, 等.基于ATR-FTIR结合CARS算法测定蜂王浆中水溶性蛋白质和总糖含量[J].中国食品学报, 2021, 21(11):183-190.
GUO C, WU W C, CHEN D, et al.Determination of water-soluble protein and total sugar in royal jelly based on attenuated total reflection mid-infrared spectroscopy and CARS algorithm[J].Journal of Chinese Institute of Food Science and Technology, 2021, 21(11):183-190.
[16] 杜尚广. 基于近红外光谱技术快速评价芸苔属种子活力[D].南昌:南昌大学, 2015.
DU S G.Rapid assessment of seed vigor of Brassica based on near on infrared spectroscopy[D].Nanchang:Nanchang University, 2015.
[17] 张亚坤. 基于多尺度高光谱成像的大豆养分检测方法研究[D].哈尔滨:东北农业大学, 2018.
ZHANG Y K.Research on detection method of soybean nutrient based on multiscale hyperspectral imaging technology[D].Harbin:Northeast Agricultural University, 2018.
[18] WU D, SUN D W.Potential of time series-hyperspectral imaging (TS-HSI) for non-invasive determination of microbial spoilage of salmon flesh[J].Talanta, 2013, 111:39-46.
[19] CHEN L J, YANG Z L, HAN L J.A review on the use of near-infrared spectroscopy for analyzing feed protein materials[J].Applied Spectroscopy Reviews, 2013, 48(7):509-522.
[20] SWATLAND H J.Internal Fresnel reflectance from meat microstructure in relation to pork paleness and pH[J].Food Research International, 1997, 30(8):565-570.
[21] SIVERTSEN A H, KIMIYA T, HEIA K.Automatic freshness assessment of cod (Gadus morhua) fillets by Vis/Nir spectroscopy[J].Journal of Food Engineering, 2011, 103(3):317-323.
[22] 程玮玮. 低温保藏猪肉有机生物大分子光谱成像规律与快速检测方法研究[D].广州:华南理工大学, 2018.
CHENG W W.Study on spectral imaging characteristics and rapid detection method of pork meat organic macromolecules preserved at low temperature[D].Guangzhou, South China University of Technology, 2018.
[23] KAMRUZZAMAN M, ELMASRY G, SUN D W, et al.Prediction of some quality attributes of lamb meat using near-infrared hyperspectral imaging and multivariate analysis[J].Analytica Chimica Acta, 2012, 714:57-67.
[24] 张伏生, 汪鸿, 韩悌, 等.基于偏最小二乘回归分析的短期负荷预测[J].电网技术, 2003, 27(3):36-40.
ZHANG F S, WANG H, HAN T, et al.Short-term load forecasting based on partial least-squares regression[J].Power System Technology, 2003, 27(3):36-40.
[25] 陈雅, 姜凯译, 李耀翔, 等.基于近红外的PE包装蓝莓新鲜度无损检测[J].包装工程, 2022, 43(7):1-10.
CHEN Y, JIANG K Y, LI Y X, et al.Nondestructive detection of freshness of PE packaged blueberries based on NIR[J].Packaging Engineering, 2022, 43(7):1-10.
[26] 龙家美. 基于高光谱和多光谱成像技术的干制红枣水分、VC和总糖含量无损检测[D].南京:南京农业大学, 2021.
LONG J M.Non-destructive detection of moisture, VC and total sugar content of dried jujube based on hyperspectral and multispectral imaging technology[D].Nanjing:Nanjing Agricultural University, 2021.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司