食品与发酵工业 >

2021 , Vol. 47 >Issue 5: 186 - 192

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.025054

分析与检测

基于高光谱成像技术的酿酒高粱品种分类

  • 孙婷 ,
  • 田建平 ,
  • 胡新军 ,
  • 罗惠波 ,
  • 黄丹 ,
  • 黄浩平
展开
  • 1(四川轻化工大学 机械工程学院, 四川 宜宾, 644000)
    2(四川轻化工大学 生物工程学院, 四川 宜宾, 644000)
    3(酿酒生物技术及应用四川省重点实验室, 四川 宜宾, 644000)
硕士研究生(胡新军讲师为通讯作者, E-mail:xjhu@suse.edu.cn)

收稿日期: 2020-07-14

  修回日期: 2020-09-16

  网络出版日期: 2021-03-31

基金资助

四川省科技厅重大科技专项项目(2018GZ0112);四川轻化工大学研究生创新基金项目(y2019003);自贡市重点科技计划项目(2018CXJD06)

收起

Classification of liquor sorghum varieties based on hyperspectral imaging technology

  • SUN Ting ,
  • TIAN Jianping ,
  • HU Xinjun ,
  • LUO Huibo ,
  • HUANG Dan ,
  • HUANG Haoping
Expand
  • 1(College of Mechanical Engineering, Sichuan University of Science & Engineering, Yibin 644000, China)
    2(College of Biotechnology Engineering, Sichuan University of Science & Engineering, Yibin 644000, China)
    3(Sichuan Key Laboratory of Brewing Biotechnology and Application, Yibin 644000, China)

Received date: 2020-07-14

  Revised date: 2020-09-16

  Online published: 2021-03-31

Fold

摘要

针对不同品种酿酒高粱分类难、分类精度有待提高等问题, 提出了一种结合光谱和图像信息的高光谱成像技术从而对酿酒高粱进行分类的方法。通过采集11类共550个高粱样本的高光谱数据, 运用连续投影算法从多元散射校正预处理后光谱中筛选出48个特征波长, 再提取图像的灰度共生矩阵作为图像特征, 利用纹理特征、全光谱、特征光谱及其结合图像特征分别建立支持向量机、偏最小二乘判别和极限学习机分类模型, 最后再采集220个未参与建模样本对所建模型进行外部验证。结果表明, 基于特征光谱结合纹理特征建立的支持向量机模型效果最佳, 训练集和测试集的识别率分别为96%和95.3%, 验证集的识别率达到91.8%, 高于单一光谱数据建模效果, 说明光谱和图像信息结合可以提高酿酒高粱的分类识别率。该方法为高粱品种的高精度分类和不同酿酒原料的快速无损检测提供了可行的方法。

关键词： 高粱; 高光谱; 特征波长; 灰度共生矩阵; 机器学习

本文引用格式

孙婷 , 田建平 , 胡新军 , 罗惠波 , 黄丹 , 黄浩平 . 基于高光谱成像技术的酿酒高粱品种分类[J]. 食品与发酵工业, 2021 , 47(5) : 186 -192 . DOI: 10.13995/j.cnki.11-1802/ts.025054

Abstract

In order to solve the problem in classifying liquor sorghum, and to improve the classification accuracy, a method using hyperspectral imaging technology combined with spectral and image information was proposed. The hyperspectral data of 550 sorghum samples from 11 classes were collected, 48 characteristic wavelengths were selected using the successive projection algorithm from the pre-processed spectrum of multiple scattering correction, and then the gray level co-occurrence matrix were extracted as image feature, the support vector machine (SVM), partial least squares discriminant analysis and extreme learning machine classification models were established using the texture feature, full spectrum, feature spectrum and their combined image feature. Finally, 220 non-participating modeling samples were collected for external verification of the built models. The results showed that SVM model based on the feature spectrum combined with texture feature had the best effect, and the recognition rate of the training set and testing set was 96% and 95.3%, respectively. The recognition rate of the verification set was 91.8%, which was higher than the modeling effect of single spectral data, the effects indicated that the combination of spectral and image information could improve classification recognition rate of liquor sorghum. A feasible method was developed for high-precision classification of sorghum varieties and rapid non-destructive testing of different brewing materials.

Key words： sorghum; hyperspectral; characteristic wavelength; grey level co-occurrence matrix; machine learning

参考文献

[1] 蒋兰.酿酒高粱淀粉含量测定及性质研究[D].重庆:重庆大学, 2013.
JIANG L.Determination of starch content and properties of brewing sorghum[D].Chongqing:Chongqing University, 2013.
[2] 毛祥, 温雪瓶, 黄丹, 等.5种常用酿酒高粱的主要成分及淀粉特性差异分析[J].中国酿造, 2020, 39(3):57-62.
MAO X, WEN X P, HUANG D, et al.Analysis of the difference in main components and starch characteristics of five commonly used brewing sorghum[J].China Brewing, 2020, 39(3):57-62.
[3] 买书魁, 杨洋, 赵小波, 等.基于NIR的白酒酿酒高粱中关键指标的定量分析[J].食品科技, 2019, 44(2):301-307.
MAI S K, YANG Y, ZHAO X B, et al.Quantitative analysis of key indexes in liquor sorghum based on NIR[J].Food Science and Technology, 2019, 44(2):301-307.
[4] 王勇生, 李洁, 王博, 等.基于近红外光谱扫描技术对高粱中粗脂肪、粗纤维、粗灰分含量的测定方法研究[J].中国粮油学报, 2020, 35(3):181-185.
WANG Y S, LI J, WANG B, et al.Study on the determination method of crude fat, crude fiber and crude ash content in sorghum based on near infrared spectroscopy scanning technology[J].Journal of the Chinese Cereals and Oils Association, 2020, 35(3):181-185.
[5] 邵琦, 陈云浩, 杨淑婷, 等.基于随机森林算法的玉米品种高光谱图像鉴别[J].地理与地理信息科学, 2019, 35(5):34-39.
SHAO Q, CHEN Y H, YANG S T, et al.Hyperspectral image identification of corn varieties based on random forest algorithm[J].Geography and Geo-Information Science, 2019, 35(5):34-39.
[6] 赵鹏, 唐艳慧, 李振宇.支持向量机复合核函数的高光谱显微成像木材树种分类[J].光谱学与光谱分析, 2019, 39(12):3 776-3 782.
ZHAO P, TANG Y H, LI Z Y.Classification of wood tree species based on hyperspectral microscopic imaging of support vector machine compound kernel function[J].Spectroscopy and Spectral Analysis, 2019, 39(12):3 776-3 782.
[7] ZHANG J, DAI L M, CHENG F.Classification of frozen corn seeds using hyperspectral VIS/NIR reflectance imaging[J].Molecules, 2019, 24(1).
[8] 刘翠玲, 林珑, 于重重, 等.基于深度学习的花生高光谱图像分类方法研究[J].计算机仿真, 2020, 37(3):189-192;283.
LIU C L, LIN N, YU C C, et al.Research on peanut hyperspectral image classification method based on deep learning[J].Computer Simulation, 2020, 37(3):189-192;283.
[9] 韩仲志, 刘杰.高光谱亚像元分解预测花生中的黄曲霉毒素B1[J].中国食品学报, 2020, 20(3):244-250.
HAN Z Z, LIU J.Prediction of aflatoxin B_1 in peanut by hyperspectral subpixel decomposition[J].Chinese Journal of Food Science, 2020, 20(3):244-250.
[10] 邹志勇, 吴向伟, 陈永明, 等.低温冷冻和机械损伤条件下马铃薯高光谱图像特征响应特性研究[J].光谱学与光谱分析, 2019, 39(11):3 571-3 578.
ZOU Z Y, WU X W, CHEN Y M, et al.Characteristic response characteristics of potato hyperspectral image under low temperature freezing and mechanical damage[J].Spectroscopy and Spectral Analysis, 2019, 39(11):3 571-3 578.
[11] 王彩霞, 王松磊, 贺晓光, 等.高光谱技术融合图像信息的牛肉品种识别方法研究[J].光谱学与光谱分析, 2020, 40(3):911-916.
WANG C X, WANG S L, HE X G, et al.Research on beef variety recognition method based on hyperspectral technology and image information[J].Spectroscopy and Spectral Analysis, 2020, 40(3):911-916.
[12] HUANG M, HE C J, ZHU Q B, et al.Maize seed variety classification using the integration of spectral and image features combined with feature transformation based on hyperspectral imaging[J].Applied Sciences, 2016, 6(6):183-194.
[13] 邢素霞, 王九清, 陈思, 等.基于K-means-RBF的鸡肉品质分类方法研究[J].食品科学技术学报, 2018, 36(4):93-99.
XING S X, WANG J Q, CHEN S, et al.Research on chicken quality classification method based on K-means-RBF[J].Journal of Food Science and Technology, 2018, 36(4):93-99.
[14] 刘爽, 谭鑫, 刘成玉, 等.高光谱数据处理算法的小麦赤霉病籽粒识别[J].光谱学与光谱分析, 2019, 39(11):3 540-3 546.
LIU S, TAN X, LIU C Y, et al.Recognition of wheat head blight based on hyperspectral data processing algorithm[J].Spectroscopy and Spectral Analysis, 2019, 39(11):3 540-3 546.
[15] HE P H, WU Y, WANG J J, et al.Detection of mites Tyrophagus putrescentiae and Cheyletus eruditus in flour using hyperspectral imaging system coupled with chemometrics[J].Journal of Food Process Engineering, 2020, 43(6).
[16] 黄平捷, 李宇涵, 俞巧君, 等.基于SPA和多分类SVM的紫外-可见光光谱饮用水有机污染物判别方法研究[J].光谱学与光谱分析, 2020, 40(7):2 267-2 272.
HUANG P J, LI Y H, YU Q J, et al.Research on the identification method of drinking water organic pollutants based on SPA and multi-class SVM[J].Spectroscopy and Spectral Analysis, 2020, 40(7):2 267-2 272.
[17] 宋镇, 姬长英, 张波.基于光谱与图像信息的杏鲍菇多糖含量检测[J].华南农业大学学报, 2019, 40(3):104-110.
SONG Z, JI C Y, ZHANG B.Pleurotus eryngii polysaccharide content detection based on spectrum and image information[J].Journal of South China Agricultural University, 2019, 40(3):104-110.
[18] 芦兵, 孙俊, 杨宁, 等.基于荧光透射谱和高光谱图像纹理的茶叶病害预测研究[J].光谱学与光谱分析, 2019, 39(8):2 515-2 521.
LU B, SUN J, YANG N, et al.Research on tea disease prediction based on fluorescence transmission spectrum and hyperspectral image texture[J].Spectroscopy and Spectral Analysis, 2019, 39(8):2 515-2 521.
[19] 赵刘, 齐海军, 金秀, 等.面向大米分类的高光谱特征波长提取方法[J].江苏农业科学, 2018, 46(17):235-238.
ZHAO L, QI H J, JIN X, et al.Hyperspectral feature wavelength extraction method for rice classification[J].Jiangsu Agricultural Sciences, 2018, 46(17):235-238.
[20] 黄亚伟, 张令, 王若兰.新陈玉米的拉曼光谱快速判别研究[J].现代食品科技, 2014, 30(12):149-152.
HUANG Y W, ZHANG L, WANG R L.Raman spectroscopy of newly-aged corn rapid discrimination research[J].Modern Food Science and Technology, 2014, 30(12):149-152.
[21] 高彤, 吴静珠, 林珑, 等.基于NIR和PLS-DA法的东北大米产地快速溯源方法研究[J].中国粮油学报, 2019, 34(7):114-117;124.
GAO T, WU J S, LIN L, et al.Research on the rapid traceability method of rice producing area in northeast china based on NIR and PLS-DA method[J].Journal of the Chinese Cereals and Oils Association, 2019, 34(7):114-117;124.
[22] 吴亚榕, 王欢, 李键红.适用于数据分类的极限学习机优化算法[J].软件导刊, 2019, 18(6):10-13;17.
WU Y R, WANG H, LI J H.Optimization algorithm of extreme learning machine suitable for data classification[J].Software Guide, 2019, 18(6):10-13;17.
[23] 陆银标.基于极限学习机的苹果叶片病害识别[D].杨凌:西北农林科技大学, 2019.
LU Y B.Apple leaf disease identification based on extreme learning machine[D].Yangling:Northwest A&F University, 2019.
[24] 董小栋, 郭培源, 徐盼, 等.融合高光谱和图像深度特征的腊肉分类与检索算法研究[J].食品工业科技, 2018, 39(23):255-260;266.
DONG X D, GUO P Y, XU P, et al.Research on bacon classification and retrieval algorithm combining hyperspectral and image depth features[J].Food Industry Science and Technology, 2018, 39(23):255-260;266.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司