食品与发酵工业 >

2022 , Vol. 48 >Issue 16: 270 - 276

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.029223

分析与检测

基于近红外光谱技术联合极限学习机的蓝莓贮藏品质定量模型建立

  • 朱金艳 ,
  • 朱玉杰 ,
  • 冯国红 ,
  • 曾明飞 ,
  • 刘思岐
展开
  • (东北林业大学 工程技术学院,黑龙江 哈尔滨,150040)
硕士研究生(朱玉杰教授和冯国红副教授为共同通信作者,E-mail:zhuyujie004@126.com;fgh_1980@126.com)

收稿日期: 2021-09-07

  修回日期: 2021-10-19

  网络出版日期: 2022-09-16

基金资助

中央高校基本科研业务费专项资金项目(2572020BL01);黑龙江省自然科学基金项目(LH2020C050)

收起

Establishment of quantitative models for blueberry storage quality based on near infrared spectroscopy combined with extreme learning machine

  • ZHU Jinyan ,
  • ZHU Yujie ,
  • FENG Guohong ,
  • ZENG Mingfei ,
  • LIU Siqi
Expand
  • College of Engineering and Technology, Northeast Forestry University, Harbin 150040, China

Received date: 2021-09-07

  Revised date: 2021-10-19

  Online published: 2022-09-16

Fold

摘要

采用近红外光谱技术联合极限学习机(extreme learning machine,ELM)方法建立蓝莓贮藏品质的定量检测模型,实现对蓝莓果实的可溶性固形物、维生素C和花青素含量的快速无损检测,以期为鲜食蓝莓低温贮藏期间的在线品质检测提供技术参考。利用LabSpec 5000光谱仪采集5个不同贮藏时间共150组蓝莓样本的近红外光谱,通过基于联合X/Y的异常样本识别和剔除方法筛选异常样本,使用联合X-Y距离样本集划分方法对样本集进行划分。通过对比分析标准正态变换、多元散射校正、一阶导数等预处理方法对模型性能的影响,确定蓝莓3个成分各自最优预处理方法,采用联合区间偏最小二乘算法(synergy interval partial least squares,SiPLS)选择出特征波段,将其作为输入建立ELM定量分析模型,并将模型结果与偏最小二乘回归进行对比分析。结果表明,蓝莓果实的可溶性固形物、维生素C和花青素含量最优ELM模型的校正集相关系数分别为0.920 5、0.908 7、0.942 1;验证集相关系数为0.882 6、0.897 2、0.869 3;校正集均方根误差为0.766 4、0.695 4、1.671 0;验证集均方根误差为0.539 7、0.624 3、2.041 4。该研究利用全光谱的1/5~2/5的变量就能达到比原始变量所建模型更好的性能,与传统的偏最小二乘回归模型相比,该文建立的ELM模型精度有明显提高,表明SiPLS-ELM结合近红外光谱技术在蓝莓成分的在线无损检测方面具有很大潜力。

关键词： 近红外光谱; 极限学习机; 蓝莓; 联合区间偏最小二乘; 无损检测

本文引用格式

朱金艳 , 朱玉杰 , 冯国红 , 曾明飞 , 刘思岐 . 基于近红外光谱技术联合极限学习机的蓝莓贮藏品质定量模型建立[J]. 食品与发酵工业, 2022 , 48(16) : 270 -276 . DOI: 10.13995/j.cnki.11-1802/ts.029223

Abstract

Near infrared spectroscopy technology joint extreme learning machine (ELM) method was used to establish a quantitative detection model of blueberry storage quality, so as to achieve the rapid non-destructive detection of the soluble solids, vitamin C and anthocyanin of blueberry fruit and provide technical reference for online quality testing during the low temperature storage of fresh blueberries. The LabSpec 5000 spectrometer was used to collect near infrared spectra of 150 blueberry samples in 5 different storage times, the abnormal samples were screened by combined X/Y anomaly sample identification and rejection method, and the sample set was divided by the joint X-Y distance sample set division method. By comparing and analyzing the effect of pretreatment methods such as standard normal variate, multiplicative scatter correction, first-order derivative on model performance, the optimal pretreatment method of each of the three components of blueberries was determined. The characteristic band was selected by the synergy interval partial least squares algorithm (SiPLS), and used as input to establish the ELM quantitative analysis model. The results of the ELM model were compared with the partial least-squares regression (PLSR). The results showed that the correction set correlation coefficients of the soluble solids, vitamin C and anthocyanin content of blueberries were 0.920 5, 0.908 7 and 0.942 1 respectively. The verification set correlation coefficients were 0.882 6, 0.897 2, and 0.869 3 respectively. The root mean square error of correction (RMSEC) were 0.766 4, 0.695 4, 1.671 0, respectively. The root mean square error of prediction (RMSEP) were 0.539 7, 0.624 3, 2.041 4, respectively. In this study, 1/5-2/5 variables of the whole spectrum were used to achieve better performance than the model established by the original variables. Compared with the traditional PLSR model, the accuracy of the ELM model established in this paper was significantly improved, indicating that SiPLS-ELM combined with near infrared spectroscopy has great potential in the online nondestructive detection of blueberry components.

Key words： near infrared spectroscopy; extreme learning machine; blueberry; synergy interval partial least squares; non-destructive testing

参考文献

[1] 谢国芳, 刘娜, 卢丹, 等.不同时间采收的贵州主栽蓝莓果实品质的综合评价[J].经济林研究, 2020, 38(2):209-214;240.
XIE G F, LIU N, LU D, et al.Comprehensive evaluation of fruit quality of main blueberry cultivars harvested at different dates in Guizhou[J].Non-Wood Forest Research, 2020, 38(2):209-214;240.
[2] 路辉, 彭彬倩, 冯晓宇, 等.大米直链淀粉、蛋白质、脂肪、水分含量的近红外光谱检测模型优化[J].中国稻米, 2020, 26(6):55-59;63.
LU H, PENG B Q, FENG X Y, et al.Model optimization for determination of amylose, protein, fat and moisture content in rice by near-infrared spectroscopy[J].China Rice, 2020, 26(6):55-59;63.
[3] 赵思梦, 于宏威, 高冠勇, 等.花生蛋白组分及其亚基含量近红外分析检测方法[J].光谱学与光谱分析, 2021, 41(3):912-917.
ZHAO S M, YU H W, GAO G Y, et al.Rapid determination of protein components and their subunits in peanut based on near infrared technology[J].Spectroscopy and Spectral Analysis, 2021, 41(3):912-917.
[4] 张鹏, 陈帅帅, 李江阔, 等.采用近红外光谱进行采后苹果品种及货架期定性判别[J].食品与发酵工业, 2019, 45(19):200-205.
ZHANG P, CHEN S S, LI J K, et al.Near-infrared spectroscopy for qualitative identification of postharvest apple varieties and shelf life[J].Food and Fermentation Industries, 2019, 45(19):200-205.
[5] 雍婧姣, 王霞, 石思佳, 等.基于R语言的近红外光谱对甘草中指标成分定量分析[J].中国实验方剂学杂志, 2019, 25(9):176-181.
YONG J J, WANG X, SHI S J, et al.Quantitative analysis of index components in glycyrrhizae Radix et khizoma by near infrared spectroscopy based on R software[J].Chinese Journal of Experimental Traditional Medical Formulae, 2019, 25(9):176-181.
[6] 杨晓辉, 张正华.近红外分析仪测定汽油辛烷值方法的建立[J].石油化工技术与经济, 2016, 32(6):14-18.
YANG X H, ZHANG Z H.Determination of octane number of gasoline with near infrared spectroscopy[J].Technology-Economics in Petrochemicals, 2016, 32(6):14-18.
[7] 王保义, 赵硕, 张少敏.基于云计算和极限学习机的分布式电力负荷预测算法[J].电网技术,2014,38(2):526-531.
WANG B Y, ZHAO S, ZHANG S M.A distributed load forecasting algorithm based on cloud computing and extreme learning machine[J].Power System Technology, 2014, 38(2):526-531.
[8] OUYANG T H, WANG C W, YU Z J, et al.Quantitative analysis of gas phase IR spectra based on extreme learning machine regression model[J].Sensors (Basel, Switzerland), 2019, 19(24):5 535.
[9] 罗一甲, 祝赫, 李潇涵, 等.赤霞珠酿酒葡萄总酚含量的近红外光谱定量分析[J].光谱学与光谱分析, 2021, 41(7):2 036-2 042.
LUO Y J, ZHU H, LI X H, et al.Quantitative analysis of total phenol content in cabernet sauvignon grape based on near-infrared spectroscopy[J].Spectroscopy and Spectral Analysis, 2021, 41(7):2 036-2 042.
[10] BUREAU S, RUIZ D, REICH M, et al.Rapid and non-destructive analysis of apricot fruit quality using FT-near-infrared spectroscopy[J].Food Chemistry, 2009, 113(4):1 323-1 328.
[11] CAYUELA J A.Prediction of intact nectarine quality using a Vis/NIR portable spectrometer[J].International Journal of Postharvest Technology and Innovation, 2011, 2(2):131.
[12] 开启余. 紫外分光光度法测定VC银翘片中维生素C含量[J].福建分析测试, 2015, 24(3):35-37.
KAI Q Y.Ultraviolet spectrophotometric method determination of the content of the vitamin C in the VC fructus forsythiae flakes[J].Fujian Analysis & Testing, 2015, 24(3):35-37.
[13] 付晓伟, 吴晓, 姜莉莉, 等.紫外分光光度法测定水果中维生素C含量[J].山东化工, 2020, 49(24):102-103;105.
FU X W, WU X, JIANG L L, et al.Determination of vitamin C in fruits by UV spectrophotometry[J].Shandong Chemical Industry, 2020, 49(24):102-103;105.
[14] 王姗姗. 蓝莓可溶性固形物、总酚和花青素近红外光谱检测技术研究[D].北京:北京林业大学, 2012.
WANG S S.Research on near infrared spectroscopy for rapid measurements of SSC, total phenols, and anthocyanins in blueberry[D].Beijing:Beijing Forestry University, 2012.
[15] 曹建康, 姜微波, 赵玉梅.果蔬采后生理生化实验指导[M].北京:中国轻工业出版社, 2007.
CAO J K, JIANG W B, ZHAO Y M.Experimental Guidance of Postharvest Physiology and Biochemistry of Fruits and Vegetables[M].Beijing:China Light Industry Press, 2007.
[16] 尹宝全, 史银雪, 孙瑞志, 等.近红外多组分分析中异常样本识别方法[J].农业机械学报, 2015, 46(S1):122-127.
YIN B Q, SHI Y X, SUN R Z, et al.Outlier samples detection method for NIR multicomponent analysis[J].Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(S1):122-127.
[17] 张娟. 基于SPXY-WT-CARS算法的草莓糖度近红外光谱检测研究[J].食品与发酵科技, 2020, 56(6):136-139;142.
ZHANG J.Research on the detection of strawberry sugar content by NIR based on SPXY-WT-CARS algorithm[J].Food and Fermentation Sciences & Technology, 2020, 56(6):136-139;142.
[18] 王世芳, 韩平, 崔广禄, 等.SPXY算法的西瓜可溶性固形物近红外光谱检测[J].光谱学与光谱分析, 2019, 39(3):738-742.
WANG S F, HAN P, CUI G L, et al.The NIR detection research of soluble solid content in watermelon based on SPXY algorithm[J].Spectroscopy and Spectral Analysis, 2019, 39(3):738-742.
[19] 赵荣军, 邢新婷, 吕建雄, 等.粗皮桉木材力学性质的近红外光谱方法预测[J].林业科学, 2012, 48(6):106-111.
ZHAO R J, XING X T, LYU J X, et al.Estimation of wood mechanical properties of Eucalyptus pellita by near infrared spectroscopy[J].Scientia Silvae Sinicae, 2012, 48(6):106-111.
[20] 易克传, 曾其良, 张新伟, 等.基于SiPLS算法的近红外光谱检测梨可溶性固形物含量[J].光谱实验室, 2013, 30(1):68-72.
YI K C, ZENG Q L, ZHANG X W, et al.Measurement of soluble solids content in pear by NIR spectroscopy based on synergy interval partial least-squares[J].Chinese Journal of Spectroscopy Laboratory, 2013, 30(1):68-72.
[21] 倪鸿飞, 斯乐婷, 黄家鹏, 等.近红外光谱结合遗传算法优化的极限学习机实现银杏叶纯化过程有效成分快速测定[J].中国中药杂志, 2021, 46(1):110-117.
NI H F, SI L T, HUANG J P, et al.Rapid determination of active components in Ginkgo biloba leaves by near infrared spectroscopy combined with genetic algorithm joint extreme learning machine[J].China Journal of Chinese Materia Medica, 2021,46(1):110-117.
[22] 丁姣, 蔡建荣, 张海东, 等.近红外结合Si-ELM检测食醋品质指标[J].食品与机械, 2012, 28(1):93-96.
DING J, CAI J R, ZHANG H D, et al.Measurement of quality index in vinegar using near infrared(NIR) combined with Si-ELM[J].Food & Machinery, 2012, 28(1):93-96.
[23] 关婷予, 黄咏梅, 林敏, 等.大米蛋白粉多组分含量近红外光谱快速检测[J].中国粮油学报, 2021, 36(6):136-142.
GUAN T Y, HUANG Y M, LIN M, et al.Determination of multi-component constituents of rice protein powder rapidly by near infrared spectroscopy[J].Journal of the Chinese Cereals and Oils Association, 2021, 36(6):136-142.
[24] 刘小路, 薛璐, 鲁晓翔, 等.近红外光谱技术快速无损检测蓝莓总黄酮、花青素的研究[J].食品工业科技, 2015, 36(16):58-61;67.
LIU X L, XUE L, LU X X, et al.Fast non-destructive testing of total flavonoids and anthocyanins in blueberries by near-infrared spectroscope[J].Science and Technology of Food Industry, 2015, 36(16):58-61;67.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司