Phân loại bệnh lá lúa bằng phương pháp chưng cất kiến thức

  • Saiful Niaz Md. Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh
  • Hassan Shabonty Tasmia Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh
  • Dewan Md. Farid Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh
  • Lucky Talukder Tarin Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh
  • Ashfaqur Rahman Saad Md. Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh
  • Dewan Md. Farid Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh
  • Huu - Hoa Nguyen College of Information and Communication Technology, Can Tho University, 92000, Can Tho, Vietnam

Abstract

Lúa đóng vai trò thiết yếu đối với an ninh lương thực toàn cầu, đặc biệt tại các quốc gia châu Á, nơi lúa là thực phẩm chính và nguồn kinh tế quan trọng. Việc trồng lúa đối mặt với nhiều thách thức, bao gồm các loại bệnh có thể gây ảnh hưởng nghiêm trọng đến năng suất và chất lượng mùa màng. Các phương pháp phát hiện bệnh truyền thống thường chậm, tốn nhiều công sức và dễ xảy ra sai sót, do đó cần thiết phải có các giải pháp tự động hóa. Nghiên cứu này tìm hiểu việc áp dụng phương pháp chưng cất kiến thức– phương pháp chuyển giao kiến thức từ các mô hình phức tạp sang các mô hình đơn giản hơn– nhằm nhận diện bệnh trong cây lúa. Chúng tôi đã tối giản hóa kiến thức từ các mô hình phức tạp bằng cách huấn luyện chuyên sâu, tạo ra một mô hình hiệu quả và gọn nhẹ. DenseNet121 được sử dụng làm mô hình giáo viên, và mô hình học sinh được trang bị hệ thống đường dẫn kép để giảm thiểu tình trạng quá khớp. Cả hai mô hình đều trải qua quá trình huấn luyện nhận thức lượng tử (QAT). Mô hình hợp tác giữa giáo viên và học sinh đạt độ chính xác 98,47%, với số lượng tham số ít hơn đáng kể (5,9 triệu) so với mô hình chỉ dùng giáo viên. Cuối cùng, toàn bộ mô hình được lượng tử hóa từ số thực 32-bit sang số nguyên 8-bit, tạo ra một phương pháp hiệu quả hơn để tách biệt các đặc tính nhỏ và phức tạp một cách chính xác. Mô hình nhỏ gọn này lý tưởng cho các thiết bị di động trong môi trường hạn chế tài nguyên, giúp nông dân nhận diện bệnh nhanh chóng, hỗ trợ quản lý cây
trồng tốt hơn và tăng cường an ninh lương thực toàn cầu.

Author Biographies

Saiful Niaz Md., Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh

Md. Saiful Niaz holds a Bachelor of Science in Computer Science and Engineering from United International University. His academic and research pursuits focus on machine learning, deep learning, computer vision, and natural language processing, fields in which he has contributed to several peer-reviewed conference publications. Mr. Niaz is also credentialed in networking and cybersecurity through Cisco’s CCNA program, underscoring his technical proficiency across programming, web development, and robotics. His back ground reflects a commitment to advancing knowledge in applied computing and intelligent systems.

Hassan Shabonty Tasmia, Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh

Tasmia Hassan Shabonty is currently completing her BSc in Computer Science at United International University, Bangladesh. She was awarded the Study in Canada Scholarship 2022 and spent a semester at Lakehead University, Thunder Bay, Canada. Her research interests span artificial intelligence, machine learning, deep learning, and human-computer interaction (HCI). She is particularly focused on leveraging AI and ML techniques to enhance user interactions and learning experiences in digital platforms. In her recent work, she explores how advanced AI models can be applied within HCI frameworks to create more intuitive and adaptive systems that support user needs.

Dewan Md. Farid, Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh

Dewan Md. Farid is a Professor of Computer Science and Engineering at United International University. He is an IEEE Senior Member and Member ACM. Prof. He holds a PhD in Computer Science and Engineering from Jahangirnagar University, Bangladesh in 2012. Part of his PhD research has been done at ERIC Laboratory,
University Lumière Lyon 2, France by Erasmus-Mundus ECW eLink PhD Exchange Program. His PhD was fully funded by Ministry of Science and Information and Communication Technology, Government of the People’s Republic of Bangladesh and European Union (EU) eLink project. Prof. Farid has published 142 peer-reviewed scientific articles, including 33 highly esteemed journals like Expert Systems with Applications, IEEE Access,
Journal of Theoretical Biology, Journal of Neuroscience Methods, Bioinformatics, Scientific Reports (Nature), Proteins and so on in the field of Machine Learning, Data Mining and Big Data.

Lucky Talukder Tarin, Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh

Lucky Talukder Tarin is a Computer Science and Engineering graduate from United International University, is committed to advancing technology through impactful research and innovative solutions. Her recent
publications highlight her dedication to creating tools that promote transparency, sustainability, and improved outcomes across various sectors.

Ashfaqur Rahman Saad Md., Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh

Md. Ashfaqur Rahman Saad is currently completing his BSc in Computer Science at United International University, Bangladesh. His research interests include machine learning, deep learning, artificial intelligence, computer networks, and cyber security. Saad is particularly interested in the intersection of AI and cybersecurity, focusing on how machine learning models can be leveraged to enhance network security and protect against emerging cyber threats. His work aims to develop intelligent systems capable of adaptive threat detection and response to strengthen digital infrastructure.

Dewan Md. Farid, Department of Computer Science and Engineering, United International University United City, Madani Avenue, Badda, Dhaka 1212, Bangladesh

Dewan Md. Farid is a Professor of Computer Science and Engineering at United International University. He is an IEEE Senior Member and Member ACM. Prof. He holds a PhD in Computer Science and Engineering from Jahangirnagar University, Bangladesh in 2012. Part of his PhD research has been done at ERIC Laboratory,
University Lumière Lyon 2, France by Erasmus-Mundus ECW eLink PhD Exchange Program. His PhD was fully funded by Ministry of Science and Information and Communication Technology, Government of the People’s Republic of Bangladesh and European Union (EU) eLink project. Prof. Farid has published 142 peer-reviewed scientific articles, including 33 highly esteemed journals like Expert Systems with Applications, IEEE Access,
Journal of Theoretical Biology, Journal of Neuroscience Methods, Bioinformatics, Scientific Reports (Nature), Proteins and so on in the field of Machine Learning, Data Mining and Big Data.

Huu - Hoa Nguyen, College of Information and Communication Technology, Can Tho University, 92000, Can Tho, Vietnam

Huu-Hoa Nguyen received his Engineering Degree in Computer Science from Can Tho University, Vietnam. He earned his MSc in Information Systems from HAN University, the Netherlands, and his PhD in Informatics from Lyon University, France. Dr. Nguyen is currently a senior lecturer at the College of Information and Communication Technology, Can Tho University, Vietnam. His research interests span a wide range of computer science topics, including artificial intelligence, machine learning, data mining, knowledge
management systems, computer networks, and cybersecurity. He has led multiple international research projects, receiving funding from prestigious sources such as the Horizon-2020 European Commission and the Newton Fund.

References

S. Sen, R. Chakraborty, and P. Kalita, “Rice-not just a staple food: A comprehensive review on its phytochemicals and therapeutic potential,” Trends in Food Science & Technology, vol. 97, pp. 265–285, 2020.

“What is bangladesh? the kernel importance description,” of rice accessed: 03-17. [Online]. Available: https://typeset.io/questions/ what-is-the-importance-of-rice-in-bangladesh-4ibrf6r1us

R. Reinke, S. Kim, and B. Kim, “Developing japonica rice introgression lines with multiple resistance genes for

brown planthopper, bacterial blight, rice blast, and rice stripe virus using molecular breeding,” Molecular Genetics and Genomics, vol. 293, no. 6, pp. 1565–1575, 2018.

M. Agrawal and S. Agrawal, “Rice plant diseases detection & classification using deep learning models: A systematic review,” Journal of Critical Reviews, vol. 7, no. 11, pp. 4376–4390, 2020.

J. Chen, D. Zhang, Y. Nanehkaran, and D. Li, “Detection of rice plant diseases based on deep transfer learning,” Journal of the Science of Food and Agriculture, vol. 100, no. 7, pp. 3246–3256, 2020.

S. Jadhav, V. Udupi, and S. Patil, “Identification of plant diseases using convolutional neural networks,” International Journal of Information Technology, vol. 13, no. 6, pp. 24612470, 2021.

T. Daniya and S. Vigneshwari, “Deep neural network for disease detection in rice plant using the texture and deep features,” The Computer Journal, vol. 65, no. 7, pp. 1812 1825, 2022.

J. Li, X. Zhao, H. Xu, L. Zhang, B. Xie, J. Yan, L. Zhang, D. Fan, and L. Li, “An interpretable high-accuracy method

for rice disease detection based on multisource data and transfer learning,” Plants, vol. 12, no. 18, p. 3273, 2023.

A. Sharma, A. Jain, P. Gupta, and V. Chowdary, “Machine learning applications for precision agriculture: A comprehensive review,” IEEE Access, vol. 9, pp. 4843–4873, 2020.

G. Hinton, O. Vinyals, and J. Dean, “Distilling the knowledge in a neural network,” 2015. [Online]. Available:

https://arxiv.org/abs/1503.02531

A. Musa, M. Hassan, M. Hamada, and F. Aliyu, “Low power deep learning model for plant disease detection for smart-hydroponics using knowledge distillation techniques,” Journal of Low Power Electronics and Applications, vol. 12, no. 2, p. 24, 2022.

Y. Hu, G. Liu, Z. Chen, J. Liu, and J. Guo, “Lightweight one-stage maize leaf disease detection model with knowledge distillation,” Agriculture, vol. 13, no. 9, p. 1664, 2023.

B. Petchiammal, Kiruba and P. Murugan, Arjunan, “Paddy doctor: A visual image dataset for automated paddy disease classification and benchmarking,” in CODS-COMAD ’23. Association for Computing Machinery, New York, NY, USA, 2023, pp. 203–207.

S. Lamba, A. Baliyan, and V. Kukreja, “A novel gcl hybrid classification model for paddy diseases,” International Journal of Information Technology, 2022.

W. Chen, J. Chen, R. Duan, Y. Fang, Q. Ruan, and D. Zhang, “Ms-dnet: A mobile neural network for

plant disease identification,” Computers and Electronics in Agriculture, vol. 199, p. 107175, 2022. [Online].

Available: https://www.sciencedirect.com/science/article/pii/ S0168169922004926

S. Nalini, N. Krishnaraj, J. Thangaiyan, K. Vinothkumar, A. Britto, K. Subramaniam, and C. Bharatiraja, “Paddy leaf disease detection using an optimized deep neural network,” Computers, Materials & Continua, vol. 680, pp. 1117–1128, 2021.

M. Niaz, L. Tarin, T. Shabonty, M. Saad, and D. Farid, “Paddy leaf disease detection employing transfer learning,” in 2023 26th International Conference on Computer and Information Technology (ICCIT), 2023, pp. 1–6. in 2024

C. Simhadri and H. Kondaveeti, “Automatic recognition of rice leaf diseases using transfer learning,” Agronomy, vol. 13, no. 4, 2023. [Online]. Available: https://www.mdpi. com/2073-4395/13/4/961

R. Narmadha, N. Sengottaiyan, and R.J.K., “Deep transfer learning based rice plant disease detection model,” Intelligent Automation & Soft Computing, vol. 31, no. 2, pp. 1257–1271, 2022. [Online]. Available: http://www. techscience.com/iasc/v31n2/44553

V. Gautam, N. Trivedi, A. Singh, H. Mohamed, I. Noya, P. Kaur, and N. Goyal, “A transfer learning-based

artificial intelligence model for leaf disease assessment,” Sustainability, vol. 14, no. 20, 2022. [Online]. Available:

https://www.mdpi.com/2071-1050/14/20/13610

M. Islam, M. Shuvo, M. Shamsojjaman, S. Hasan, M. Hossain, and T. Khatun, “An automated convolutional neural network based approach for paddy leaf disease detection,” International Journal of Advanced Computer Science and Applications, vol. 12, no. 1, 2021.

A. Ghofrani and R. Mahdian Toroghi, “Knowledge distillation in plant disease recognition,” Neural Computing and Applications, vol. 34, no. 17, pp. 14287–14296, 2022. [Online]. Available: https: //doi.org/10.1007/s00521-021-06882-y

Y. Hu, G. Liu, Z. Chen, J. Liu, and J. Guo, “Lightweight one-stage maize leaf disease detection model with knowledge distillation,” Agriculture, vol. 13, no. 9, 2023. [Online]. Available: https://www.mdpi.com/2077-0472/13/9/1664

A. Musa, M. Hassan, M. Hamada, and F. Aliyu, “Low power deep learning model for plant disease detection for smart-hydroponics using knowledge distillation techniques,” Journal of Low Power Electronics and Applications, vol. 12, no. 2, 2022. [Online]. Available: https://www.mdpi.com/ 2079-9268/12/2/24

“Rice leaf disease detection obj dataset,” visited on 2024 03-18. [Online]. Available: https://universe.roboflow.com/ project-khcjh/rice-leaf-disease-detection-obj

“Rice leaf diseases dataset kernel description,” accessed: 2023-08-14. [Online]. Available: https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases

“Rice diseases image dataset,” accessed: 2023-08.14. [Online]. Available: https://www.kaggle.com/datasets/

minhhuy2810/rice-diseases-imagedataset/data

Published
2024-11-06