食品与发酵工业 >

2022 , Vol. 48 >Issue 12: 255 - 260

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.028482

分析与检测

基于近红外光谱和支持向量机回归参数调优的羊肉含水量检测

  • 张立欣 ,
  • 杨翠芳 ,
  • 张晓 ,
  • 张楠楠 ,
  • 王亚明
展开
  • 1(塔里木大学 信息工程学院,新疆 阿拉尔,843300)
    2(南京理工大学 理学院,江苏 南京,210094)
第一作者:张立欣(博士、副教授)和杨翠芳(讲师)为共同第一作者(王亚明高级实验师为通信作者,E-mail:279508585@qq.com)

收稿日期: 2021-07-01

  修回日期: 2021-08-19

  网络出版日期: 2022-07-15

基金资助

国家自然科学基金(31960503);国家自然科学基金(61662064);塔里木大学校长基金(TDZKSS202006);塔里木大学校长基金(TDZKQN201709)

收起

Detection and parameter optimization of moisture content in mutton based on near infrared spectroscopy and SVR

  • ZHANG Lixin ,
  • YANG Cuifang ,
  • ZHANG Xiao ,
  • ZHANG Nannan ,
  • WANG Yaming
Expand
  • 1(College of Information Engineering, Tarim University, Alaer 843300,China)
    2(School of science, Nanjing University of Science and Technology, Nanjing 210094,China)

Received date: 2021-07-01

  Revised date: 2021-08-19

  Online published: 2022-07-15

Fold

摘要

羊肉中的水分含量直接影响着其加工、贮藏和口感,因此对其水分含量的检测具有十分重要的意义。在900~1 700 nm的波长范围内采集南疆羊肉的光谱数据,采用一阶导数(first derivative,1-DER)、标准正态变换(standard normal transformation,SNV)、多元散射校正(multivariate scatter correction,MSC)、小波变换(wave transformation,WT)、SG平滑变换(Savitzky Golag smooth transformation,SG)、傅里叶变换(Fourier transform,FT)对原始光谱数据进行预处理。分别采用连续投影算法(successive projection algorithm, SPA)和竞争自适应重加权算法(competitive adaptive reweighted sampling, CARS)进行光谱特征选取,建立偏最小二乘回归(partial least squares regression,PLS)和支持向量机回归(support vector regression,SVR)模型对羊肉水分含量进行预测。结果显示,采用1-DER-CARS-SVR模型,选取参数c为0.701 1,g为0.088 4时,预测效果最佳,测试集的均方误差为1.216 2,拟合优度为0.739 5。研究结果为研发羊肉水分含量的无损检测装置提供理论参考。

关键词： 近红外光谱; 连续投影算法; 竞争自适应重加权算法; 偏最小二乘回归; 支持向量基回归

本文引用格式

张立欣 , 杨翠芳 , 张晓 , 张楠楠 , 王亚明 . 基于近红外光谱和支持向量机回归参数调优的羊肉含水量检测[J]. 食品与发酵工业, 2022 , 48(12) : 255 -260 . DOI: 10.13995/j.cnki.11-1802/ts.028482

Abstract

Detection of moisture content of mutton is very important because it directly affects the processing, storage, and taste. The near-infrared spectrum data of mutton in south Xinjiang were collected in the wavelength range of 900-1 700 nm. Noise among the spectrum data was removed to enhance the accuracy and robustness of the model by six pretreated methods, including first derivative, standard normal variate, multiplicative scatter correction, wavelet transforms, SG smoothing transform, and Fourier transform. Then, successive projections algorithm and competitive adaptive reweighted sampling algorithm were used to select characteristic variables respectively. Finally, partial least squares regression (PLS) and support vector regression (SVR) models were established to predict the moisture content of mutton. Results showed that the 1-DER-CARS-SVR model with parameters c was 0.701 1 and g was 0.088 4 which had the best performance. The mean square error of the test set was 1.216 2, the goodness of fit was 0.739 5. This provides theoretical references for the further study on the nondestructive testing detection device of moisture content of mutton in south Xinjiang.

Key words： near infrared spectrum; successive projection algorithm; competitive adaptive reweighted sampling; partial least squares regression; support vector regression

参考文献

[1] 白京,李家鹏,邹昊,等.近红外特征光谱定量检测羊肉卷中猪肉掺假比例[J].食品科学,2019,40(2):287-292.
BAI J, LI J P, ZOU H, et al.Quantitative detection of pork in adulterated mutton rolls based on near infrared spectroscopy[J].Food Science,2019,40(2):287-292.
[2] MABOOD F, BOQUÉ R, ALKINDI A Y, et al.Fast detection and quantification of pork meat in other meats by reflectance FT-NIR spectroscopy and multivariate analysis[J].Meat Science, 2020, 163:108084.
[3] 王靖,丁佳兴,郭中华,等.基于近红外高光谱成像技术的宁夏羊肉产地鉴别[J].食品工业科技,2018,39(2):250-254;260.
WANG J, DING J X, GUO Z H, et al.Identification of geographical origins of mutton in Ningxia based on the near infrared hyperspectral imaging technique[J].Science and Technology of Food Industry,2018,39(2):250-254;260.
[4] 王彩霞, 何智武, 吴龙国, 等.高光谱技术结合特征波长筛选的牛肉品种多波段识别[J].发光学报,2019,40(4):520-527.
WANG C X, HE Z W, WU L G, et al.Multi-bands recognition of beef breeds with hyperspectral technology combined with characteristic wavelengths selection methods[J].Chinese Journal of Luminescence, 2019,40(4):520-527.
[5] 刘贵珊,张翀, 樊奈昀, 等.IVISSA算法冷鲜滩羊肉嫩度的高光谱模型优化[J].光谱学与光谱分析,2020,40(8):2 558-2 563.
LIU G S, ZHANG C, FAN N Y, et al.Hyperspectral model optimization for tenderness of chilled tan-sheep mutton based on IVISSA[J].Spectroscopy and Spectral Analysis, 2020, 40(8):2 558-2 563.
[6] YU H D, ZUO S M,XIA G H, et al.Rapid and nondestructive freshness determination of Tilapia fillets by a portable near-infrared spectrometer combined with chemometrics methods[J].Food Analytical Methods,2020,13(10):1 918-1 928.
[7] CHENG L J, LIU G S, HE J G, et al.Non-destructive assessment of the myoglobin content of Tan sheep using hyperspectral imaging[J].Meat Science,2020,167:107988.
[8] 邱园园,朱荣光,黄昆鹏,等.真空包装冷却羊肉pH值和颜色的近红外快速无损检测[J].新疆农业科学,2017,54(11):2 092-2 099.
QIU Y Y, ZHU R G, HUANG K P, et al.Rapid and nondestructive detection of the pH valueand and Luminosity for vacuum packed chilled mutton by using near infrared reflectance spectroscopy technique[J].Xinjiang Agricultural Sciences,2017,54(11):2 092-2 099.
[9] 袁凯,张志勇,席前,等.3步混合变量选择策略在鸡肉近红外水分检测中的应用[J].食品与机械,2020,36(9):72-76;81.
YUAN K, ZHANG Z Y, XI Q, et al.Research on the application of three-step hybrid variable selection strategy in chicken moisture detection by near infrared[J].Food & Machinery, 2020,36(9):72-76;81.
[10] 陈杰,姚娜,吕海芳,张晓.近红外光谱在小尾寒羊羊肉水分预测中的应用[J].食品科技,2021,46(1):134-138.
CHEN J, YAO N, LYU H F, et al.Application of near infrared spectroscopy in water prediction of small-tail han sheep[J].Food Science and Technology,2021,46(1):134-138.
[11] 张立欣,李文华,王顺,等.基于ELM算法的羊肉水分含量的快速无损检测[J].数学的实践与认识,2020,50(8):94-102.
ZHANG L X, LI W H, WANG S, et al.Fast non-destructive detection of moisture content in mutton based on ELM algorithm[J].Mathematics in Practice and Theory, 2020,50(8):94-102.
[12] 第五鹏瑶,卞希慧,王姿方,等.光谱预处理方法选择研究[J].光谱学与光谱分析,2019,39(9):2 800-2 806.
DIWU P Y, BIAN X G, WANG Z F, et al.Study on the selection of spectral preprocessing methods[J].Spectroscopy and Spectral Analysis, 2019,39(9):2 800-2 806.
[13] 杨倩圆,谢定,郑瑞娜,等.淀粉液化液葡萄糖当量值的近红外快速检测[J].食品工业科技,2019,40(9):248-252.
YANG Q Y, XIE D, ZHENG R N, et al.Fast detection of starch liquefying solution′s dextrose equivalent value by near infrared spectroscopy[J].Science and Technology of Food Industry,2019,40(9):248-252.
[14] FENG C H, MAKINO Y.Colour analysis in sausages stuffed in modified casings with different storage days using hyperspectral imaging—A feasibility study[J].Food Control, 2020, 111:107047.
[15] PANG L, WANG J H, MEN S, et al.Hyperspectral imaging coupled with multivariate methods for seed vitality estimation and forecast for Quercus variabilis[J].Spectrochimica Acta Part A Molecular and Biomolecular Spectroscopy, 2021, 245:118888.
[16] 吴鹏,宋海燕,杨威,等..基于近红外光谱的鸡蛋产地溯源[J].食品工业科技,2020,41(22):227-231.
WU P, SONG H Y, YANG W, et al.The origin of eggs based on near infrared spectroscopy[J].Science and Technology of Food Industry,2020,41(22):227-231.
[17] 王亚轩,谭峰,辛元明,等.大米拉曼光谱不同预处理方法的相近产地鉴别研究[J].光谱学与光谱分析,2021,41(2):565-571.
WANG Y X, TAN F,XIN Y M, et al.Identification of rice from similar areas with different pretreatment methods of Raman spectrum[J].Spectrocopy and Spectral Analysis,2021,41(2):565-571.
[18] 宋相中,唐果,张录达,等.近红外光谱分析中的变量选择算法研究进展[J].光谱学与光谱分析,2017,37(4):1 048-1 052.
SONG X Z, TANG G, ZHANG L D, et al.Research advance of variable selection algorithms in near infrared spectroscopy analysis[J].Spectrocopy and Spectral Analysis,2017,37(4):1 048-1 052.
[19] 曹晓兰,邓梦洁,崔国贤.高光谱结合主成分分析的苎麻品种识别[J].光谱学与光谱分析,2019,39(6):1 905-1 908.
CAO X L, DENG M J, CUI G X.Identifying ramie variety by combining the hyperspectral technology with the principal component analysis[J].Spectrocopy and Spectral Analysis,2019,39(6):1 905-1 908.
[20] CHEN S M, HU T T, LUO L H, et al.Rapid estimation of leaf nitrogen content in apple-trees based on canopy hyperspectral reflectance using multivariate methods[J].Infrared Physics & Technology, 2020, 111:103542.
[21] ZHANG D Y, YANG Y, CHEN G, et al.Nondestructive evaluation of soluble solids content in tomato with different stage by using Vis/NIR technology and multivariate algorithms[J].Spectrochimica Acta Part A Molecular and Biomolecular Spectroscopy, 2021, 248:119139.
[22] 段宇飞,王巧华.基于特征选择与特征提取融合的鸡蛋新鲜度光谱快速检测模型优化[J].食品科学,2020,41(12):273-278.
DUAN Y F, WANG Q H.Optimization of a predictive model for rapid detection of egg freshness using visible near-infrared spectra based on combination of feature selection and feature extraction[J].Food Science,2020,41(12):273-278.
[23] 刘秀英,余俊茹,王世华.光谱特征变量和BP神经网络构建油用牡丹种子含水率估算模型[J].农业工程学报,2020,36(22):308-315.
LIU X Y, YU J R, WANG S H.Estimation of moisture content in peony seed oil using spectral characteristic variables and BP neural network[J].Transactions of the Chinese Society of Agricultural Engineering, 2020,36(22):308-315.
[24] FANG Y M, YANG F, ZHOU Z, et al.Hyperspectral wavelength selection and integration for bruise detection of Korla pears[J].Journal of Spectroscopy,2019, 2019:6715247.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司