Objective:To solve the problem of low accuracy of tobacco grading. Methods:An improved tobacco leaf grading model based on convolutional neural network was proposed. According to the VGG16 network structure, the network model was built in a custom way. The traditional convolution was replaced by the hole convolution, which increased the image receptive field while avoiding. The loss of image features was changed, and the activation function was changed to Leaky_relu. The data distribution was corrected, and the hard saturation problem of the ReLU function was solved. 41 levels of tobacco leaf pictures were used for testing. Results:The grading accuracy rate of the test algorithm was 95.89%, which was 10.46% higher than the traditional SVM algorithm, and 7.87% higher than the classic VGG16 algorithm. The loss rate finally converged to 0.13. Conclusion:Compared with the original model and traditional feature extraction Methods, this algorithm has improved the accuracy of tobacco leaf classification.

Publication Date


First Page


Last Page





[1] 郭强.基于图像处理与神经网络的烟叶分级研究[D].长沙:中南大学,2013:12-16.GUO Qiang.Research on tobacco leaf classification based on image processing and neural Network[D].Changsha:Central South University,2013:12-16.
[2] 杨帆,申金媛.基于BPSO和SVM的烤烟烟叶图像特征选择方法研究[J].湖北农业科学,2015,54(2):449-452.YANG Fan,SHEN Jin-yuan.Research on image feature selection method of flue-cured tobacco based on BPSO and SVM[J].Hubei Agricultural Sciences,2015,54(2):449-452.
[3] 苏明秋.基于烟叶图像的参数精确提取和分级识别系统研究与实现[D].成都:电子科技大学,2020:1-5.SU Ming-qiu.Research and implementation of accurate parameter extraction and classification recognition system based on tobacco leaf image[D].Chengdu:University of Electronic Science and Technology of China,2020:1-5.
[4] 轻工业部烟草工业科学研究所.《烤烟》国家标准制订的依据和说明[S].北京:技术标准出版社,1982:1-8.Tobacco Industry Scientific Research Institute of the Ministry of Light Industry.Basis and explanation for the formulation of the national standard of "flue-cured tobacco"[S].Beijing:Technical Standard Press,1982:1-8.
[5] 招启柏,冯柱安,周兴华,等.禄丰基地烤烟生态环境和烟叶质量分析[J].中国农业气象,2012,33(1):98-103.ZHAO Qi-bai,FENG Zhu-an,ZHOU Xing-hua,et al.Analysis of the ecological environment and tobacco leaf quality of flue-cured tobacco in Lufeng base[J].Chinese Journal of Agricultural Meteorology,2012,33(1):98-103.
[6] ZHANG Fan,ZHANG Xin-hong.Classification and quality evaluation of tobacco leaves based on image processing and fuzzy comprehensive evaluation[J].Sensors,2011,11(3):2 369-2 384.
[7] 姚学练,贺福强,平安,等.基于PCA-GA-SVM的烟叶分级方法[J].烟草科技,2018,51(12):98-105.YAO Xue-lian,HE Fu-qiang,PING An,et al.Tobacco leaf classification method based on PCA-GA-SVM[J].Tobacco Science & Technology,2018,51(12):98-105.
[8] 杨双艳,杨紫刚,张四伟,等.基于近红外光谱和PSO-SVM算法的烟叶自动分级方法[J].贵州农业科学,2018,46(12):141-144.YANG Shuang-yan,YANG Zi-gang,ZHANG Si-wei,et al.Automatic classification method of tobacco leaves based on near infrared spectroscopy and PSO-SVM algorithm[J].Guizhou Agricultural Sciences,2018,46(12):141-144.
[9] 邓晨曦.基于机器视觉的烟叶自动分级方法研究[J].科技创新导报,2020,17(12):39-40.DENG Chen-xi.Research on Automatic Tobacco classification method based on machine vision[J].Science & Technology Innovation Herald,2020,17(12):39-40.
[10] LI Rui-dong,ZHANG Xiao-bing,LI Ke-qiang,et al.Nondestruct-ive and rapid grading of tobacco leaves by use of a hand-held near-infrared spectrometer,based on a particle swarm optimiza-tion-extreme learning machine algorithm[J].Spectroscopy Letters,2020,53(9):685-691.
[11] 曾祥云.一个基于深度学习的烤烟分级系统的设计与实现[D].南京:东南大学,2017:5-7.ZENG Xiang-yun.Design and implementation of a flue-cured tobacco grading system based on deep learning[D].Nanjing:Southeast University,2017:5-7.
[12] 雒慧心.基于深度学习的烤烟分选算法研究[D].北京:北京交通大学,2019:2-6.LUO Hui-xin.Research on sorting algorithm of flue-cured tobacco based on deep learning[D].Beijing:Beijing Jiaotong University,2019:2-6.
[13] 王士鑫,云利军,叶志霞,等.一种基于卷积神经网络的烟叶分级处理算法[J].云南民族大学学报(自然科学版),2020,29(1):65-69.WANG Shi-xin,YUN Li-jun,YE Zhi-xia,et al.Atobacco leaf classification processing algorithm based on convolutional neural network[J].Journal of Yunnan Nationalities University(Natural Science Edition),2020,29(1):65-69.
[14] 赵春雷,苏瑶,寇霄腾,等.预处理方法对再造烟叶中烟梗品质的影响[J].食品与机械,2017,33(5):211-215.ZHAO Chun-lei,SU Yao,KOU Xiao-teng,et al.Effects of pretreatment Methods on the quality of tobacco stems in reconstituted tobacco[J].Food & Machinery,2017,33(5):211-215.
[15] LIU Jian-jun,SHEN Jin-yuan,SHEN Zhen-yu,et al.Grading tobacco leaves based on image processing and generalized regression neural network[C]//2012 IEEE International Conference on Intelligent Control.New York:IEEE Press,Automatic Detection and High-End Equipment,2012:89-93.
[16] MARZAN C S,RUIZ C R.Automated tobacco grading using image processing techniques and a convolutional neural network[J].International Journal of Machine Learning and Computing,2019,9(6):807.
[17] 刘帅奇,王洁,安彦玲,等.基于CNN的非下采样剪切波域多聚焦图像融合[J].郑州大学学报(工学版),2019,40(4):7.LIU Shuai-qi,WANG Jie,AN Yan-ling,et al.Non-subsampled shear wave domain multi-focus image fusion based on CNN[J].Journal of Zhengzhou University(Engineering Edition),2019,40(4):7.

Included in

Food Science Commons



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.