Abstract: Freezing damage is an important factor that causes rot in sweet potatoes.Timely detection and elimination of freezing damage of sweet potato is the key to reducing production losses.To establish a rapid and non-destructive detection method for freezing damage of sweet potato in the early stage, this study established a recognition model for the freezing damage of sweet potato in the early stage based on hyperspectral technology and chemometric methods."Gaia Sorter", a kind of hyperspectral sorter, was used to collect the hyperspectral information of the intact and frost-damaged sweet potato samples in the 900-1 700 nm band, and extract the average spectrum of the intact and frost-damaged regions of the samples to obtain 343 spectra for the intact region and 476 spectra for the frost-damaged region.The Kennard-Stone algorithm was used to select intact and frost-damaged samples in the training set and prediction set.Four preprocessing methods were used to preprocess the original spectra, and the first derivative (FD) was selected as the best preprocessing method.The competitive adaptive reweighted sampling (CARS), iteratively retains informative variables (IRIV), and CARS-SPA and IRIV-SPA combined with successive projections algorithm (SPA) were used to screen out 46, 65, 24 and 35 characteristic wavelength variables related to sweet potato freezing damage.Partial least squares (PLS) and least squares support vector machines (LS-SVM) were used to establish identification models of freezing damage in sweet potato, and it was compared with the recognition model without variable screening.The results showed that hyperspectral technology could effectively detect the freezing damage of sweet potatoes.The CARS method could effectively select useful wavelength variables.It was superior to the IRIV, CARS-SPA and IRIV-SPA algorithms.The CARS-PLS model showed fast calculation speed and the best prediction results.The recognition accuracy, sensitivity and specificity of samples in the prediction set were 98.05%, 98.84% and 97.48%, respectively.This study has achieved the identification of the characteristics of freezing damage in sweet potato, which lays a theoretical foundation for the subsequent research on online sorting and detection and the development of online testing equipment for the sweet potato quality.
许建东,张淑娟,郑小南,等. 高光谱技术结合变量选择方法的甘薯冻害检测研究[J]. 食品与发酵工业, 2021, 47(8): 197-203.
XU Jiandong,ZHANG Shujuan,ZHENG Xiaonan,et al. Study on the detection of sweet potato freezing damage based on hyperspectral technology and variable selection method[J]. Food and Fermentation Industries, 2021, 47(8): 197-203.
FERNANDEZ A O, PINATTI C A O, PERIS R M, et al.Freeze-damage detection in lemons using electrochemical impedance spectroscopy[J].Sensors, 2019, 19(18):4 051.
[2]
KIM H C, JO H N, KIM Y O.Shooting and fruiting characteristics of 'hachiya' persimmon tree affected by sub-zero temperature treatment at early budding stage[J].The Journal of the Convergence on Culture Technology, 2019, 5(4):395-399.
[3]
CHARRIER G, CHUINE I, BONHOMME M, et al.Assessing frost damages using dynamic models in walnut trees:Exposure rather than vulnerability controls frost risks[J].Plant, Cell &Environment, 2018, 41(5):1 008-1 021.
[4]
马惠玲, 王若琳, 蔡骋, 等.基于高光谱成像的苹果品种快速鉴别[J].农业机械学报, 2017, 48(4):305-312.MA H L, WANG R L, CAI C, et al.Rapid identification of apple varieties based on hyperspectral imaging[J].Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(4):305-312.
[5]
杨福芹, 冯海宽, 李振海, 等.基于可见光-近红外光谱特征参数的苹果叶片氮含量预测[J].农业机械学报, 2017, 48(9):143-151.YANG F Q, FENG H K, LI Z H, et al.Prediction for nitrogen content of apple leaves using spectral features parameters from visible and near infrared lights[J].Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(9):143-151.
[6]
冯迪, 纪建伟, 张莉, 等.基于高光谱成像提取苹果糖度与硬度最佳波长[J].发光学报, 2017, 38(6):799-806.FENG D, JI J W, ZHANG L, et al.Optimal wavelengths extraction of apple brix and firmness based on hyperspectral imaging[J].Chinese Journal of Luminescence, 2017, 38(6):799-806.
[7]
陈欣欣, 郭辰彤, 张初, 等.高光谱成像技术的库尔勒梨早期损伤可视化检测研究[J].光谱学与光谱分析, 2017, 37(1):150-155.CHEN X X, GUO C T, ZHANG C, et al.Visual detection study on early bruises of korla pear based on hyperspectral imaging technology[J].Spectroscopy and Spectral Analysis, 2017, 37(1):150-155.
[8]
李雄, 刘燕德, 欧阳爱国, 等.酥梨货架期的高光谱成像无损检测模型研究[J].光谱学与光谱分析, 2019, 39(8):2 578-2 583.LI X, LIU Y D, OUYANG A G, et al.Study on non-destructive testing model of hyperspectral imaging for shelf life of crisp pear[J].Spectroscopy and Spectral Analysis, 2019, 39(8):2 578-2 583.
[9]
潘廷跳. 基于光谱和成像技术的香梨黑斑病发病过程监测与快速检测方法研究[D].广州:华南理工大学, 2019.PAN T T.Study on pathogenetic process monitoring and rapid detection methods of pear black spot disease based on spectroscopic and spectral imaging technologies[D].Guangzhou:South China University of Technology, 2019.
[10]
朱晓琳, 李光辉, 张萌.基于CARS-MIV-SVR的库尔勒香梨可溶性固体含量预测方法[J].光谱学与光谱分析, 2019, 39(11):3 547-3 552.ZHU X L, LI G H, ZHANG M, et al.Prediction of soluble solid content of korla pears based on CARS-MIV[J].Spectroscopy and Spectral Analysis, 2019, 39(11):3 547-3 552.
[11]
姜微. 高光谱技术在马铃薯品种鉴别及品质无损检测中的应用研究[D].哈尔滨:东北农业大学, 2017.JIANG W.Study on nondestructive detection of identification varieties and quality potato using hyperspectral technology[D].Harbin:Northeast Agricultural University, 2017.
[12]
吉海彦, 任占奇, 饶震红.基于高光谱成像技术的不同产地小米判别分析[J].光谱学与光谱分析, 2019, 39(7):2 271-2 277.JI H Y, REN Z Q, RAO Z H.Discriminant analysis of millet from different origins based on hyperspectral imaging technology[J].Spectroscopy and Spectral Analysis, 2019, 39(7):2 271-2 277.
[13]
陈李品, 于繁千惠, 陶然, 等.基于高光谱成像技术预测牡蛎干制加工过程中的水分含量[J].中国食品学报, 2020, 20(7):261-268.CHEN L P, YU F Q H, TAO R, et al.Prediction of moisture content in oyster drying process based on hyperspectral imaging[J].Journal of Chinese Institute of Food Science and Technology, 2020, 20(7):261-268.
[14]
ZHANG L, SUN H, RAO Z H, et al.Hyperspectral imaging technology combined with deep forest model to identify frost-damaged rice seeds[J].Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy, 2020, 229.
[15]
WEI C W, HUANG J F, WANG X Z, et al.Hyperspectral characterization of freezing injury and its biochemical impacts in oilseed rape leaves[J].Remote Sensing of Environment, 2017, 195:56-66.
[16]
史萍, 武永峰, 胡新, 等.晚霜冻影响下冬小麦冠层光谱特征与株高要素变化关系[J].光谱学与光谱分析, 2017, 37(12):3 845-3 850.SHI P, WU Y F, HU X, et al.Relationship between hyperspectral parameters of winter wheat canopy and plant height components under late frost injury[J].Spectroscopy and Spectral Analysis, 2017, 37(12):3 845-3 850.
[17]
孟雷, 武永峰, 胡新, 等.土壤表层湿度影响下冬小麦晚霜冻害及冠层光谱检测[J].光谱学与光谱分析, 2017, 37(5):1 482-1 488.MENG L, WU Y F, HU X, et al.Using hyperspectral data for detecting late frost injury to winter wheat under different topsoil moistures[J].Spectroscopy and Spectral Analysis, 2017, 37(5):1 482-1 488.
[18]
魏传文. 基于多源数据的油菜冻害遥感机理与方法研究[D].杭州:浙江大学, 2018.WEI C W.Remote sensing mechanisms and methods of freezing injury in winter oilseed rape using multi-source data[D].Hangzhou:Zhejiang University, 2018.
[19]
POLESELLO A, GIANGIACOMO R.Application of near infrared spectrophotometry to the nondestructive analysis of foods:A review of experimental results[J].Critical Reviews in Food Science and Nutrition, 1983, 18(3):203-230.
[20]
宋华鲁, 闫银发, 宋占华, 等.利用介电参数和变量筛选建立玉米籽粒含水率无损检测模型[J].农业工程学报, 2019, 35(20):262-272.SONG H L, YAN Y F, SONG Z H, et al.Nondestructive testing model for maize grain moisture content established by screening dielectric parameters and variables[J].Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(20):262-272.
[21]
章海亮, 楚秉泉, 叶青, 等.高光谱成像技术鉴别鱼新鲜度[J].光谱学与光谱分析, 2018, 38(2):559-563.ZHANG H L, CHU B Q, YE Q, et al.Classification of fishness based on hyperspectra imaging technology[J].Spectroscopy and Spectral Analysis, 2018, 38(2):559-563.
[22]
李冠稳, 高小红, 肖能文, 等.基于sCARS-RF算法的高光谱估算土壤有机质含量[J].发光学报, 2019, 40(8):1 030-1 039.LI G W, GAO X H, XIAO N W, et al.Estimation soil organic matter contents with hyperspectra based on sCARS and RF algorithms[J].Chinese Journal of Luminescence, 2019, 40(8):1 030-1 039.
[23]
于雷, 章涛, 朱亚星, 等.基于IRIV算法优选大豆叶片高光谱特征波长变量估测SPAD值[J].农业工程学报, 2018, 34(16):148-154.YU L, ZHANG T, ZHU Y X, et al.Determination of soybean leaf SPAD value using characteristic wavelength variables preferably selected by IRIV algorithm[J].Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(16):148-154.
[24]
孙俊, 莫云南, 戴春霞, 等.基于介电特性与IRIV-GWO-SVR算法的番茄叶片含水率检测[J].农业工程学报, 2018, 34(14):188-195.SUN J, MO Y N, DAI C X, et al.Detection of moisture content of tomato leaves based on dielectric properties and IRIV-GWO-SVR algorithm[J].Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(14):188-195.
[25]
WEI L F, YUAN Z R, YU M, et al.Estimation of arsenic content in soil based on laboratory and field reflectance spectroscopy[J].Sensors (Basel, Switzerland), 2019, 19(18):3 904.
[26]
吴宜青, 孙通, 刘津, 等.基于LSSVM和CARS变量优选的食用植物油中铬含量DP-LIBS检测[J].激光与光电子学进展, 2018, 55(1):438-445.WU Y Q, SUN T, LIU J, et al.Detection of chromium content in edible vegetable oil with DP-LIBS combined with LSSVM and CARS methods[J].Laser & Optoelectronics Progress, 2018, 55(1):438-445.
[27]
高升, 王巧华.基于高光谱图像信息融合的红提糖度无损检测[J].发光学报, 2019, 40(12):1 574-1 584.GAO S, WANG Q H.Comprehensive detection of internal quality of red globe grape extract based on near infrared spectroscopy[J].Chinese Journal of Luminescence, 2019, 40(12):1 574-1 584.
[28]
孙通, 莫欣欣, 李晓珍, 等.近红外光谱技术结合变量选择方法定性检测食用植物油中的腐霉利[J].光谱学与光谱分析, 2016, 36(12):3 915-3 919.SUN T, MO X X, LI X Z, et al.Qualitative detection of procymidone in edible vegetable oils by near infrared spectroscopy and variable selection methods[J].Spectroscopy and Spectral Analysis, 2016, 36(12):3 915-3 919.
[29]
许文丽, 药林桃, 孙通, 等.基于CARS-SPA的苹果可溶性固形物可见/近红外光谱在线检测[J].食品工业科技, 2014, 35(22):61-64.XU W L, YAO L T, SUN T, et al.CARS-SPA baesd visble/near infraed spectroscopy on-line detection of apple soluble solids content[J].Science and Technology of Food Industry, 2014, 35(22):61-64.
[30]
李冠稳, 高小红, 肖能文, 等.特征变量选择和回归方法相结合的土壤有机质含量估算[J].光学学报, 2019, 39(9):361-371.LI G W, GAO X H, XIAO N W, et al.Estimation of soil organic matter content based on characteristic variable selection and regression methods[J].Acta Optica Sinica, 2019, 39(9):361-371.
2021, Vol. 47 
