运用近红外光谱技术对松子霉变的快速检测
|
摘要
采集234组有代表性的松子实验数据,对光谱数据进行求导、变量标准化(SNV)、小波变换、套索算法(LASSO)与主成分分析(PCA)方法预处理后,使用高斯过程(GP)等10种建模方法对光谱数据进行建模,分析运用近红外光谱技术快速检测松子霉变的可行性。结果表明:使用径向基核支持向量机建模的F1度量分数为0.868、使用高斯过程建模的F1度量分数为0.631、将应用主成分分析方法降维后的数据使用高斯过程建模的F1度量分数为0.933、将应用套索算法与主成分分析方法处理后的数据使用高斯过程建模的F1度量分数为1.000,实验结果验证了使用套索算法-主成分分析-高斯过程建立近红外光谱模型筛选霉变松子是可行的。
We collected 234 sets of representative pine nut experimental data, and used Gaussian process after derivation of spectral data, variable normalization(SNV), wavelet transform, lasso algorithm(LASSO) and principal component analysis(PCA). Gaussian Process and the other ten modeling methods modeled the spectral data and analyzed the feasibility of using near-infrared spectroscopy to quickly detect the mildew of pine nut. The F1-score modeled by RBF-SVM was 0.868, the F1 metric score modeled by Gaussian process was 0.631, and the data after dimension reduction using principal component analysis method was Gaussian process modeled with F1 score of 0.933. Using the lasso algorithm and principal component analysis method, the F1 metric score of Gaussian process modeling was 1. Therefore, it is feasible to use the LASSO-PCA-Gaussian Process to establish a near-infrared spectroscopy model to screen pine nut mildew.
We collected 234 sets of representative pine nut experimental data, and used Gaussian process after derivation of spectral data, variable normalization(SNV), wavelet transform, lasso algorithm(LASSO) and principal component analysis(PCA). Gaussian Process and the other ten modeling methods modeled the spectral data and analyzed the feasibility of using near-infrared spectroscopy to quickly detect the mildew of pine nut. The F1-score modeled by RBF-SVM was 0.868, the F1 metric score modeled by Gaussian process was 0.631, and the data after dimension reduction using principal component analysis method was Gaussian process modeled with F1 score of 0.933. Using the lasso algorithm and principal component analysis method, the F1 metric score of Gaussian process modeling was 1. Therefore, it is feasible to use the LASSO-PCA-Gaussian Process to establish a near-infrared spectroscopy model to screen pine nut mildew.
引文
[1] 仇逊超,曹军.近红外光谱波段优化在东北松子蛋白质定量检测中的应用[J].现代食品科技,2016,32(11):303-309.
[2] QU J H,LIU D,CHENG J H,et al.Applications of near-infrared spectroscopy in food safety evaluation and control:A review of recent research advances[J].Critical Reviews in Food Science and Nutrition,2015,55(13):1939-1954.
[3] FU X P,YING Y B.Food safety evaluation based on near infrared spectroscopy and imaging:a review[J].Critical Reviews in Food Science and Nutrition,2016,56(11):1913-1924.
[4] POREP J U,KAMMERER D R,CARLE R.On-line application of near infrared (NIR) spectroscopy in food production[J].Trends in Food Science & Technology,2015,46(2):211-230.
[5] 孙通,徐惠荣,应义斌.近红外光谱分析技术在农产品/食品品质在线无损检测中的应用研究进展[J].光谱学与光谱分析,2009,29(1):122-126.
[6] HUANG M,KIM M S,DELWICHE S R,et al.Quantitative analysis of melamine in milk powders using near-infrared hyperspectral imaging and band ratio[J].Journal of Food Engineering,2016,181:10-19.doi.org/10.1016/j.jfoodeng.2016.02.017.
[7] HUANG L,ZHAO J W,CHEN Q S,et al.Nondestructive measurement of total volatile basic nitrogen (TVB-N) in pork meat by integrating near infrared spectroscopy,computer vision and electronic nose techniques[J].Food Chemistry,2014,145:228-236.doi.org/10.1016/j.foodchem.2013.06.073.
[8] KAMRUZZAMAN M,MAKINO Y,OSHITA S,et al.Assessment of visible near-infrared hyperspectral imaging as a tool for detection of horsemeat adulteration in minced beef[J].Food and Bioprocess Technology,2015,8(5):1054-1062.
[9] PU H B,SUN D W,MA J,et al.Classification of fresh and frozen-thawed pork muscles using visible and near infrared hyperspectral imaging and textural analysis[J].Meat Science,2015,99:81-88.doi.org/10.1016/j.meatsci.2014.09.001.
[10] 刘媛媛,彭彦昆,王文秀,等.基于偏最小二乘投影的可见/近红外光谱猪肉综合品质分类[J].农业工程学报,2014,30(23):306-313.
[11] 武小红,孙俊,武斌,等.基于联合区间偏最小二乘判别分析的猪肉近红外光谱定性建模分析[J].激光与光电子学进展,2015,52(4):242-247.
[12] 蔡健荣,万新民,陈全胜.近红外光谱法快速检测猪肉中挥发性盐基氮的含量[J].光学学报,2009,29(10):2808-2812.
[13] DAI Q,CHENG J H,SUN D W,et al.Prediction of total volatile basic nitrogen contents using wavelet features from visible/near-infrared hyperspectral images of prawn (Metapenaeus ensis)[J].Food Chemistry,2016,197:257-265.doi.org/10.1016/j.foodchem.2015.10.073.
[14] 夏立娅,申世刚,刘峥颢,等.基于近红外光谱和模式识别技术鉴别大米产地的研究[J].光谱学与光谱分析,2013,33(1):102-105.
[15] 战皓,吴宏伟,张东,等.近红外光谱法测定不同产地黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷含量[J].光谱学与光谱分析,2017,37(5):1391-1396.
[16] 郭志明,黄文倩,陈全胜,等.近红外光谱的苹果内部品质在线检测模型优化[J].现代食品科技,2016,32(9):147-153.
[17] 李伟,李金龙,李卫军,等.基于机器学习的玉米单倍体近红外光谱鉴别方法研究[J].光谱学与光谱分析,2018,38(9):2763-2769.
[18] 陈克明,周志鑫,卢汉清,等.基于高斯过程的高分辨率遥感图像变化检测[J].遥感学报,2012,16(6):1192-1204.
[19] RINNAN ?,VAN DEN BERG F,Engelsen S B.Review of the most common pre-processing techniques for near-infrared spectra[J].TrAC Trends in Analytical Chemistry,2009,28(10):1201-1222.
[2] QU J H,LIU D,CHENG J H,et al.Applications of near-infrared spectroscopy in food safety evaluation and control:A review of recent research advances[J].Critical Reviews in Food Science and Nutrition,2015,55(13):1939-1954.
[3] FU X P,YING Y B.Food safety evaluation based on near infrared spectroscopy and imaging:a review[J].Critical Reviews in Food Science and Nutrition,2016,56(11):1913-1924.
[4] POREP J U,KAMMERER D R,CARLE R.On-line application of near infrared (NIR) spectroscopy in food production[J].Trends in Food Science & Technology,2015,46(2):211-230.
[5] 孙通,徐惠荣,应义斌.近红外光谱分析技术在农产品/食品品质在线无损检测中的应用研究进展[J].光谱学与光谱分析,2009,29(1):122-126.
[6] HUANG M,KIM M S,DELWICHE S R,et al.Quantitative analysis of melamine in milk powders using near-infrared hyperspectral imaging and band ratio[J].Journal of Food Engineering,2016,181:10-19.doi.org/10.1016/j.jfoodeng.2016.02.017.
[7] HUANG L,ZHAO J W,CHEN Q S,et al.Nondestructive measurement of total volatile basic nitrogen (TVB-N) in pork meat by integrating near infrared spectroscopy,computer vision and electronic nose techniques[J].Food Chemistry,2014,145:228-236.doi.org/10.1016/j.foodchem.2013.06.073.
[8] KAMRUZZAMAN M,MAKINO Y,OSHITA S,et al.Assessment of visible near-infrared hyperspectral imaging as a tool for detection of horsemeat adulteration in minced beef[J].Food and Bioprocess Technology,2015,8(5):1054-1062.
[9] PU H B,SUN D W,MA J,et al.Classification of fresh and frozen-thawed pork muscles using visible and near infrared hyperspectral imaging and textural analysis[J].Meat Science,2015,99:81-88.doi.org/10.1016/j.meatsci.2014.09.001.
[10] 刘媛媛,彭彦昆,王文秀,等.基于偏最小二乘投影的可见/近红外光谱猪肉综合品质分类[J].农业工程学报,2014,30(23):306-313.
[11] 武小红,孙俊,武斌,等.基于联合区间偏最小二乘判别分析的猪肉近红外光谱定性建模分析[J].激光与光电子学进展,2015,52(4):242-247.
[12] 蔡健荣,万新民,陈全胜.近红外光谱法快速检测猪肉中挥发性盐基氮的含量[J].光学学报,2009,29(10):2808-2812.
[13] DAI Q,CHENG J H,SUN D W,et al.Prediction of total volatile basic nitrogen contents using wavelet features from visible/near-infrared hyperspectral images of prawn (Metapenaeus ensis)[J].Food Chemistry,2016,197:257-265.doi.org/10.1016/j.foodchem.2015.10.073.
[14] 夏立娅,申世刚,刘峥颢,等.基于近红外光谱和模式识别技术鉴别大米产地的研究[J].光谱学与光谱分析,2013,33(1):102-105.
[15] 战皓,吴宏伟,张东,等.近红外光谱法测定不同产地黄芪中毛蕊异黄酮葡萄糖苷和黄芪甲苷含量[J].光谱学与光谱分析,2017,37(5):1391-1396.
[16] 郭志明,黄文倩,陈全胜,等.近红外光谱的苹果内部品质在线检测模型优化[J].现代食品科技,2016,32(9):147-153.
[17] 李伟,李金龙,李卫军,等.基于机器学习的玉米单倍体近红外光谱鉴别方法研究[J].光谱学与光谱分析,2018,38(9):2763-2769.
[18] 陈克明,周志鑫,卢汉清,等.基于高斯过程的高分辨率遥感图像变化检测[J].遥感学报,2012,16(6):1192-1204.
[19] RINNAN ?,VAN DEN BERG F,Engelsen S B.Review of the most common pre-processing techniques for near-infrared spectra[J].TrAC Trends in Analytical Chemistry,2009,28(10):1201-1222.