Moisture content monitoring of cigar leaves during drying based on a Convolutional Neural Network
More details
Hide details
Agriculture College, Sichuan Agricultural University, 611130 Chengdu, China
Sichuan Provincial Tobacco Company Dazhou Branch, 635000 Dazhou, China
Sichuan Provincial Tobacco Company Deyang Branch, 618400 Deyang, China
Sichuan Provincial Tobacco Company, 610017 Chengdu, China
Final revision date: 2023-04-17
Acceptance date: 2023-04-27
Publication date: 2023-06-24
Corresponding author
Zeng Shuhua   

Agricultural College, Sichuan Agricultural University, 四川成都, 000000, 四川成都, China
Int. Agrophys. 2023, 37(3): 225–234
  • Convolution neural network (CNN) method was proposed.
  • The CNN model was trained to learn the relationship between images and the corresponding moisture content using the extracted color, shape, and texture features as the input.
  • The results demonstrated that the estimated value of CNN agreed with the predicted value; the R2 was 0.9044, and the average accuracy was 87.34 %.
The moisture content of cigar leaves during drying is an important indicator for controlling the management of drying rooms. At present, the determination of cigar leaf moisture content is mainly dependent on traditional destructive detection methods, which are inefficient and damaging to plants. In this study, a Convolution Neural Network method consisting of digital images for monitoring the moisture content of cigar leaves during the drying process was proposed. In this study, the Convolution Neural Network model was trained to learn the relationship between the images and the corresponding moisture content using the extracted colour, shape, and texture features as input factors. In order to compare the Convolution Neural Network estimation results, a widely used traditional machine learning algorithm was applied. The results demonstrated that the estimated value of Convolution Neural Network agreed with the predicted value; the R2 was 0.9044, and the average accuracy was 87.34%. These results were better than those produced by traditional machine learning methods. The generalization test of the proposed method was conducted using varieties of cigar leaves in other drying rooms. The results showed that Convolution Neural Network is a viable method for an accurate estimation of the moisture content, the R2 was 0.8673 and the average accuracy was 86.81%. The Convolution Neural Network established by the features extracted from digital images could accurately estimate the moisture content of cigar leaves during drying and was therefore shown to be an effective monitoring tool.
This work was supported by a grant [SCYC202121] (2021-2024) from the China National Tobacco Corporation Sichuan Branch
The Authors declare they have no conflict of interest.
Azman A.A. and Ismail F.S., 2017. Convolutional neural network for optimal pineapple harvesting. Elektrika - J. Electr. Eng., 16(2), 1-4,
Duan S.-J., Song C.-P., Ma L., Shi L.-F., Wang W.-C., and Gong C.-R., 2012. Detection of moisture content in tobacco leaves during baking based on image processing. J. Northwest Agric. Forest. Univ., 40(05), 74-80,
Du H.-N., Meng L.-F., Wang S.-F., Zhang B.-H., Wang A.-H., Liu H., Li Z.-H., and Sun F.-S., 2022. Comparison of prediction models of water loss rate of tobacco leaves in intensive curing process based on machine learning. Tobacco Sci. Technol., 55 (09), 81-88,
Beghin T., Cope J.S., Remagnino P., and Barman S., 2010. Shape and texture based plant leaf classification. Advanced Concepts for Intelligent Vision Systems, 6475, 345-353,
Ding X.-L., Zhao L.-X., Zhou T.-T., Li Y.-B., Li, Huang X.-M., and Zhao Y-L., 2019. Research on wheat leaf water content based on machine vision. Cluster Comput., 22(S4), 9199-9208,
Erawan I.M.S., Handoyo W.T., and Sarwono W., 2021. Data integration of humidity sensor and image texture for water content prediction of Gracilaria sp. during sun drying. IOP Conf. Ser.: Earth Environ. Sci., 733(1), 012116,
Fan N.-B., Zhang R.-N., Lu X.-C., Song C.-P., Zou H.-H., Zhong Q., and Zou C.- M., 2020. The relationship between color and moisture content and membrane lipid peroxidation during the air curing of cigar tobacco leaves. Chin. Tob. Sci., 41(06), 96-102,
Ghosh A. and Roy P., 2022. An automated model for leaf image-based plant recognition: an optimal feature-based machine learning approach. Innov. Syst. Softw. Eng., 1-17,
Gong Z.-Y., Deng D., Sun X.-D., Liu J.-B., and Yang Y.-P., 2021. Non-destructive detection of moisture content for Ginkgo biloba fruit with terahertz spectrum and image: A preliminary study. Infrared Phys. Technol., 120, 103997,
Gornale S.-S., Patravali P.-U., and Hiremath P.-S., 2020. Automatic detection and classification of knee osteoarthritis using Hu’s invariant moments. Front. Robot. AI, 7, 591827,
Haralick R.-M., Shanmugam K., and Dinstein I.-H., 1973. Textural features for image classification. IEEE Trans. Syst. Man Cybern., 3(6), 610-621,
Herzig P., Borrmann P., Knauer U., Klück H., Kilias D., Seiffert U., Pillen K., and Maurer A., 2021. Evaluation of RGB and multispectral unmanned aerial vehicle (UAV) imagery for high-throughput phenotyping and yield prediction in barley breeding. Remote Sens., 13(14), 2670,
Huang Z., Zhu T.-T., Li Z.-Y., and Ni C., 2021. Non-destructive testing of moisture and nitrogen content in Pinus Massoniana seedling leaves with NIRS based on MS-SC-CNN. Applied Sci., 11(6), 2754,
Hu M.K., 1962. Visual pattern recognition by moment invariants. IEEE Trans. Inf. Theory, 8(2), 179-187,
ISO 6488:2021. Tobacco and tobacco products – Determination of water content – Karl Fischer method.
ISO 2881:1992. Tobacco and tobacco products – Preparation of test sample and determination of water content – Oven method.
Liang G.-Z., Dong C.-W., Hu B., Zhu H.-K., Yuan H.-B., Jiang Y.-W., and Hao G.-S., 2018. prediction of moisture content for congou black tea withering leaves using image features and nonlinear method. Sci. Rep., 8(1), 7854,
Mochizuki T. and Ito M., 1995. Classification of ultrasonic images using fuzzy reasoning and spatial smoothing effect of textural features. Electron. Comm. Jpn 3, 78(6), 62-76,
Okamura N.K., Shimomachi T., Takemasa T., and Takakura T., 2001. Nondestructive detection of water stress in tomato plants by NIR spectroscopy. Environ. Control Biol., 39(2), 75-85,
Qin J.-W., Wang L.-H., Jiang W., Wang J.-L., and Jia T.-P., 2021. Simulation of tobacco redrying and drying process based on COMSOL. Food Machinery, 37(11), 136-141,].
Sai R.B. and Neeraja S., 2022. Plant leaf disease classification and damage detection system using deep learning models. Multimed. Tools Appl., 81(17), 24021-24040,
Ting A., Yu H., Yang C.-S., Liang G.-Z., Chen J.-Y., Hu Z.-H., Hu B., and Dong C.-W., 2020. Black tea withering moisture detection method based on convolution neural network confidence. J. Food Proc. Engin., 43(7).
Shimomachi T., Takemasa T., Kurata K., and Takakura T., 2004. Nondestructive detection of water stress in tomato plants using microwave sensing. Environ. Control Biol., 42(1), 83-90,
Wang Z.-B, Li H., Zhu Y., and Xu T., 2017. Review of plant identification based on image processing. Arch. Comput. Methods Eng., 24(3), 637-654,
Wang R.-H., Feng J.-Y., and Wu W.-F., 2020. Correlation between moisture content and machine vision image characteristics of corn kernels. Int. J. Food Prop., 23(1), 319-328,
Wei K.-S., Jun B., Wang F., and Zhao K., 2022. On-line monitoring of the tobacco leaf composition during flue-curing by near-infrared spectroscopy and deep transfer learning. Anal. Lett., 55(13), 2089-2107,
Xu Y., Kou J.-M., Zhang Q., Tan S.-D., Zhu L.-H., Geng Z.-H., and Yang X.-H., 2023. Visual detection of water content range of seabuckthorn fruit based on transfer deep learning. Foods, 12(3), 1-14,
Yang C.-Z., 2021. Plant leaf recognition by integrating shape and texture features. Pattern Recognit., 112, 107809,
Ye H.-Y., Ding S.-S., Duan W.-J., Hu X., Lu R.-L., Guo W.-L., and Shi X.-D., 2022. Study on the synergistic change of morphology and water content of cigar tobacco leaves in the air curing process. Chin. J. Tob., 29(1),
Zang Y.-Z., Yao X.-D., Cao Y.-X., Niu Y.-B., Liu H., Xiao H.-W., Zheng X., Wang Q., and Zhu R.-G., 2021. Real-time detection system for moisture content and color change in jujube slices during drying process. J. Food Proc. Preserv., 45(6), e15539,
Zhang Q.-Y., Luo C., Li D.-L., and Cai W., 2020. Research progress in cigar tobacco leaf modulation and fermentation technology. Chin. J. Tob., 26 (04), 1-6,
Zhu L.-J., Zhang H., Zhuang Y.-D, Cao Y., Shen X.-C., and Fu J.-C., 2017. Determination of moisture retention of tobacco leaf by NIR technology. Tob. Sci. Technol., 50(9), 55-60,