Method for the classification of tea diseases via weighted sampling and hierarchical classification learning

Rujia Li, Weibo Qin, Yiting He, Yadong Li, Rongbiao Ji, Yehui Wu, Jiaojiao Chen, Jianping Yang

Abstract


This study proposed a weighted sampling hierarchical classification learning method based on an efficient backbone network model to address the problems of high costs, low accuracy, and time-consuming traditional tea disease recognition methods. This method enhances the feature extraction ability by conducting hierarchical classification learning based on the EfficientNet model, effectively alleviating the impact of high similarity between tea diseases on the model’s classification performance. To better solve the problem of few and unevenly distributed tea disease samples, this study introduced a weighted sampling scheme to optimize data processing, which not only alleviates the overfitting effect caused by too few sample data but also balances the probability of extracting imbalanced classification data. The experimental results show that the proposed method was significant in identifying both healthy tea leaves and four common leaf diseases of tea (tea algal spot disease, tea white spot disease, tea anthracnose disease, and tea leaf blight disease). After applying the “weighted sampling hierarchical classification learning method” to train 7 different efficient backbone networks, most of their accuracies have improved. The EfficientNet-B1 model proposed in this study achieved an accuracy rate of 99.21% after adopting this learning method, which is higher than EfficientNet-b2 (98.82%) and MobileNet-V3 (98.43%). In addition, to better apply the results of identifying tea diseases, this study developed a mini-program that operates on WeChat. Users can quickly obtain accurate identification results and corresponding disease descriptions and prevention methods through simple operations. This intelligent tool for identifying tea diseases can serve as an auxiliary tool for farmers, consumers, and related scientific researchers and has certain practical value.
Keywords: tea diseases, hierarchical classification learning, weighted sampling, classification method, EfficientNet, mini-program
DOI: 10.25165/j.ijabe.20241703.8236

Citation: Li R J, Qin W B, He Y T, Li Y D, Ji R B, Wu Y H, et al. Method for the classification of tea diseases via weighted sampling and hierarchical classification learning. Int J Agric & Biol Eng, 2024; 17(3): 211-221.

Keywords


tea diseases, hierarchical classification learning, weighted sampling, classification method, EfficientNet, mini-program

Full Text:

PDF

References


Zhao C J, Li J, Feng X. Developnt strategy of smart agriculture for 2035 in China. Stategic Study of CAE, 2021; 23（4）: 1–9. （in Chinese）

Picon A, Alvarez-Gila A, Seitz M, Ortiz-Barredo A, Echazarra J, Johannes A. Deep convolutional neural networks for mobile capture device-based crop disease classification in the wild. Computers and Electronics in Agriculture, 2019; 161: 280–290.

Su S F, Qiao Y, Rao Y. Recognition of grape leaf diseases and mobile application based on transfer learning. Transactions of the CSAE, 2021; 37（10）: 127–134. （in Chinese）

Lu Y N, Chen B C. Identification of hops pests and diseases in small samples based on attentional mechanisms. Journal of Chinese Agricultural Mechanization, 2021; 42（3）: 189–196. （in Chinese）

Li Z M, Xu J, Zheng L, Tie J, Yu S. Small sample recognition method of tea disease based on improved DenseNet. Transactions of the CSAE, 2022; 38（10）: 182–190. （in Chinese）

Jiang H H, Yang X H, Ding R R, Wang D W, Mao W H, Qiao Y L. Identification of apple leaf diseases based on improved ResNet18. Transactions of the CSAM, 2023; 54（4）: 295–303. （in Chinese）

Wang D F, Wang J. Crop disease classification with transfer learning and residual networks. Transactions of the CSAE, 2021; 37（4）: 199–207. （in Chinese）

Xu Y, Li X Z, Wu Z H, Gao Z, Liu L. Potato leaf disease recognition via residual attention network. Journal of Shandong University of Science and Technology （Natural Science Edition）, 2021; 40（2）: 76–83. （in Chinese）

Ji M M, Zhang K L, Wu Q F, Deng Z. Multi-label learning for crop leaf diseases recognition and severity estimation based on convolutional neural networks. Soft Computing: A fusion of foundations, methodologies and applications, 2020; 24（20）: 15327–15340.

Arnal Barbedo J G. Impact of dataset size and variety on the effectiveness of deep learning and transfer learning for plant disease classification. Computers and Electronics in Agriculture, 2018; 153: 46–53.

Sun K, He M J, He Z C, Liu H Y, Pi X T. EfficientNet embedded with spatial attention for recognition of multi-label fundus disease from color fundus photographs. Biomedical Signal Processing and Control, 2022; 77: 103768.

Freeman, I, Roese-Koerner L, Kummert A. Effnet: An efficient structure for convolutional neural networks. In: 2018 25th IEEE International Conference on Image Processing （ICIP）, Athens: IEEE, 2018; pp.6–10.

Mimi A, Zohura S F T, Ibrahim M, Haque R R, Farrok O, Jabid T, et al. Identifying selected diseases of leaves using deep learning and transfer learning models. Machine Graphics & Vision, 2023; 32（1）: 55–71.

Amin B, Samir R S, Tarek Y, Ahmed M, Ibrahim R, Ahmed M, et al. Brain tumor multi classification and segmentation in MRO images using deep learning. arXiv Preprint, arXiv: 2304.10039.

Pateria S, Subagdja B, Tan A, Quek C. Hierarchical reinforcement learning: A comprehensive survey. ACM Computing Surveys, 2021; 54（5）: 1–35.

Pourvali M, Meng Y, Sheng C, Du Y Z. TaxoKnow: Taxonomy as prior knowledge in the loss function of multi-class classification. arXiv Preprint arXiv: 2305.16341, 2023.

Lynch N, Mallmann-Trenn F. Learning hierarchically-structured concepts II: Overlapping concepts, and networks with feedback. In: International Colloquium on Structural Information and Communication Complexity, Cham: Springer Nature Switzerland, 2023; pp.46–86.

Mohammed R, Rawashdeh J, Abdullah M. Machine Learning with oversampling and undersampling techniques: Overview study and ecperimental results. In: 2020 11th International Conference on Information and Communication Systems （ICICS）, 2020; pp.243–248.

Drummond C, Holte R C. C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling. In: Workshop on learning from imbalanced datasets II, Washinton, 2003; 11: 1–8.

Trabucco B, Doherty K, Gurinas M, Salakhutdinov R. Effective data augmentation with diffusion models. arXiv preprint, arXiv: 2302.07944.

Ma X H, Li X R, Luo A, Zhang J Q, Li H. Galaxy image classification using hierarchical data learning with weighted sampling and label smoothing. Monthly Notices of the Royal Astronomical Society, 2023; 519（3）: 4765–4779.

Zhang W, Zhang P Y, Zhang B, Wang X X, Wang D. A collaborative transfer learning framework for cross-domain recommendation. arXiv Preprint, arXiv: 2306.16425.

Li M, Li Y L, Lin M. A review of transfer learning for named entity recognition. Journal of Computer Science and Technology, 2021; 15（2）: 206–218. （in Chinese）

Wan Z W, Liu C, Zhang M, Fu J, Wang B Y, Cheng S B, et al. Med-UniC: Unifying cross-lingual medical vision-language pre-training by diminishing bias. arXiv Preprint, arXiv: 2305.19894.




Copyright (c) 2024 International Journal of Agricultural and Biological Engineering

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

2023-2026 Copyright IJABE Editing and Publishing Office