Citation :

Sanjyot Thuse, Meena Chavan, "Jetson Nano-Powered Smart Farm Prototype for Real-Time Pest Classification Using a Vision Transformer–SVM Hybrid Model," International Journal of Electrical and Electronics Engineering, vol. 13, no. 5, pp. 62-77, 2026. Crossref, https://doi.org/10.14445/23488379/IJEEE-V13I5P106

Abstract

Agriculture is a key source of income for a large section of the world's population. It has become a necessity to make prompt and accurate pest detection for overall crop protection along with sustainable farming. Several current research projects rely on CNN architectures or cloud-based systems, which may not be suitable for agricultural fields with limited resources. To address this gap, the paper proposes a novel hybrid ViT-SVM model for real-time pest identification. This combines Vision Transformer, having global feature extraction capabilities, with Support Vector Machine, which has robust decision limits. Extensive tests were undertaken to evaluate several feature extractor-classifier mixtures. These combinations include ResNet, DenseNet, and ViT, combined using both SVM and Random Forest classifiers. To guarantee statistical reliability, F1-score, recall, accuracy, and precision were the metrics used to assess performance. Macro, weighted averages, class-wise ROC-AUC, PR-AUC, confusion matrices, McNemar tests, as well as bootstrap confidence intervals were calculated to test the performance of the models. The ViT-SVM hybrid system achieved superior performance to all other systems because it reached 96% accuracy and achieved an F1-score while maintaining equal success rates in detecting visually similar yet less common pest species. The system was developed as a model that ran on a Jetson Nano human-operated robotic system known as AgroPestBot, which used an Android interface for real-time pest identification at field sites. The method delivers an effective edge-AI solution that combines advanced machine learning capabilities with agricultural technology that can be used in the field to deliver farmers instant, precise pest detection results. The study demonstrates how lightweight hybrid models can be applied to resource-limited environments while establishing base technologies for upcoming automated systems and improved, precise agricultural methods.

Keywords

Deep Learning, Jetson Nano, Pest Classification, Support Vector Machine (SVM), Vision Transformer (VIT).

References

1. K. Gireesan, From Agricultural Co-operatives to Farmer Producer Companies: Analysing the Transition of Co-Operativism in India, Routledge, 2022.
   [CrossRef] [Google Scholar] [Publisher Link]
2. Alok K. Sharma, A. Rajawat, and V. Gupta, “An Analytical Study of Contribution of Agriculture Sector in Growth of Indian Economy,” Jai Maa Saraswati Gyandayni An International Multidisciplinary e-Journal, vol. 7, no. 1, pp. 08-18, 2021.
   [CrossRef] [Publisher Link]
3. S.D. Talekar, and Yonas Dubale, Economic Reforms and Agricultural Growth in India: Issues and Challenges, Economics, Agricultural and Food Sciences, 2020. [Online]. Available: https://www.semanticscholar.org/paper/Economic-Reforms-and-Agricultural-Growth-in-India%3A-Talekar-Dubale/a3abc93b7b17858e8f99eec065cffa9de13cf5f2
4. J. L. Quaresma, The Influence of Agriculture for a Country, Communication and Digital Media, vol. 1, no. 1, 2020. [Online]. Available: https://comopolo.com/index.php/comdig/article/view/2 
5. P.S. Soumia et al., Entomopathogenic Microbes for Sustainable Crop Protection: Future Perspectives, Advances in Sustainable Agriculture and Farming Systems, Springer, Singapore, pp. 469-497, 2021.
   [CrossRef] [Google Scholar] [Publisher Link]
6. Mifftha Yaseen et al., Insect Pest Infestation During Storage of Cereal Grains, Pulses and Oilseeds, Health and Safety Aspects of Food Processing Technologies, Springer, Cham, pp. 209-234, 2019,
   [CrossRef] [Google Scholar] [Publisher Link]
7. Amelia Apriyuni et al., “Analysis of the Use of PPE on the Health Risks of Farmers Spraying Pesticides,” Journal of Educational Innovation and Public Health, vol. 2, no. 3, pp. 94-114, 2024.
   [CrossRef] [Google Scholar] [Publisher Link]
8. Mayang Dwi Octavia, and Susilawati Susilawati, “Analysis of the Use of Personal Protective Equipment on the Health Status of Pesticide Spraying Farmers,” Healthy People: Journal of Public Health, vol. 2, no. 3, pp. 328-337, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
9. Christos A. Damalas, and Spyridon D. Koutroubas, “Farmers’ Exposure to Pesticides: Toxicity Types and Ways of Prevention,” Toxics, vol. 4, no. 1, pp. 1-10, 2016.
   [CrossRef] [Google Scholar] [Publisher Link]
10. B. Prasath, and M. Akila, “IoT-based Pest Detection and Classification Using Deep Features with Enhanced Deep Learning Strategies,” Engineering Applications of Artificial Intelligence, vol. 121, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
11. Sandhya Devi Ramiah Subburaj et al., “Efficient Pest Detection through Advanced Machine Learning Technique,” Current Agriculture Research Journal, vol. 12, no. 3, pp. 1127-1134, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
12. Karla Zayood, and Rama Asad Nadweh, “Automated Insect Detection and Classification Using Pelican Optimization Algorithm with Deep Learning on Internet of Enabled Agricultural Sector,” International Journal of Advances in Applied Computational Intelligence, vol. 7, no. 1, pp. 50-62, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
13. Asaf Raza et al., “Automatic Classification and Detection of Insect Pests Using Deep Transfer Learning,” 2024 International Conference on Frontiers of Information Technology (FIT), Islamabad, Pakistan, pp. 1-6, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
14. R. Prabha, and K. Selvan, “Modified RESNET50 with Attention Module for Detection and Classification of Pests in Vegetable Crops,” Journal of Advanced Research in Applied Sciences and Engineering Technology, vol. 61, no. 2, pp. 150-169, 2026.
    [CrossRef] [Google Scholar] [Publisher Link]
15. S. Senthil Pandi et al., “A Deep Learning Framework with Attention Mechanism for Accurate Detection and Classification of Crop Pest Management,” 2024 4^th International Conference on Computer, Communication, Control & Information Technology (C3IT), Hooghly, India, pp. 1-6, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
16. P. Venkatasaichandrakanth, and M. Iyapparaja, “GNViT-An Enhanced Image-Based Groundnut Pest Classification Using Vision Transformer (ViT) Model,” PLOS ONE, vol. 19, no. 3, pp. 1-23, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
17. M. Xie, and N. Ye, “Multi-Scale and Multi-Factor ViT Attention Model for Classification and Detection of Pest and Disease in Agriculture,” Applied Sciences, vol. 14, no. 13, pp. 1-14, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
18. Gabriel Savio de Lima Mota et al., “Classifying Pests in Crop Images Using Deep Learning,” Proceedings of the 18^th Workshop on Computer Vision (WVC), pp. 1-6, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
19. P. Venkatasaichandrakant and M. Iyapparaja, “Pest Detection and Classification in Peanut Crops Using CNN, MFO, and EViTA Algorithms,” IEEE Access, vol. 11, pp. 54045-54057, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
20. Nikita Agarwal et al., “An Improved Deep Learning Model Implementation for Pest Species Detection,” International Conference on Artificial Intelligence: Towards Sustainable Intelligence, Pune, India, pp. 119-131, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
21. Agus Kurniawan, Introduction to NVIDIA Jetson Nano, IoT Projects with NVIDIA Jetson Nano, Apress, Berkeley, CA, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
22. NVIDIA, Get Started with Jetson Nano Developer Kit, NVIDIA Developer, 2026. [Online]. Available: https://developer.nvidia.com/embedded/learn/get-started-jetson-nano-devkit.
23. NVIDIA, TensorRT Open-Source Software, 2019. [Online]. Available: https://github.com/NVIDIA/TensorRT
24. Gayatri Pattnaik, and K. Parvathi, “Automatic Detection and Classification of Tomato Pests Using Support Vector Machine Based on HOG and LBP Feature Extraction Technique,” Progress in Advanced Computing and Intelligent Engineering, vol. 2, pp. 49-55, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
25. Gayatri Pattnaik, and Kodimala Parvathy, “Machine Learning-Based Approaches for Tomato Pest Classification,” TELKOMNIKA Telecommunication Computing Electronics and Control, vol. 20, no. 2, pp. 321-328, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
26. Abdul Amir Abdullah Karim, and Rafal Ali Sameer, “Image Classification Using Bag of Visual Words (BoVW),” Al-Nahrain Journal of Science, vol. 21, no. 4, pp. 76-82, 2018.
    [CrossRef] [Google Scholar] [Publisher Link]
27. T. Ojala, M. Pietikäinen, and T. Maenpaa, “Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 7, pp. 971-987, 2002.
    [CrossRef] [Google Scholar] [Publisher Link]
28. Herbert Bay, Tinne Tuytelaars, and Luc Van Gool, “SURF: Speeded Up Robust Features,” European Conference on Computer Vision, Graz, Austria, pp. 404-417, 2006.
    [CrossRef] [Google Scholar] [Publisher Link]
29. S. Murakami, H. Homma, and T. Koike, “Detection of Small Pests on Vegetable Leaves Using GLCM,” American Society of Agricultural and Biological Engineers, 2005.
    [CrossRef] [Google Scholar] [Publisher Link]
30. Tao Liu et al., “Detection of Aphids in Wheat Fields Using a Computer Vision Technique,” Biosystems Engineering, vol. 141, pp. 82-93, 2016.
    [CrossRef] [Google Scholar] [Publisher Link]
31. Arti Prasad et al., “A Detailed Survey on Awareness, Knowledge and Practice of Pesticides Used Against Various Vegetables, Fruits and Cereal Crops Grown in and Around Udaipur Region of South Rajasthan, India,” Bulletin of Pure and Applied Sciences- Zoology, vol. 42A, no. 1, pp. 43-63, 2023.
    [Google Scholar] [Publisher Link]
32. Santosh Kumar Sahu, and Manish Pandey, “An Optimal Hybrid Multiclass SVM For Plant Leaf Disease Detection Using Spatial Fuzzy C-Means Model,” Expert Systems with Applications, vol. 214, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
33. Thierry Tchokogoué et al., “A Robust Segmentation Method Combined with Classification Algorithms for Field-Based Diagnosis of Maize Plant Phytosanitary State,” Journal of Intelligent Systems, vol. 33, no. 1, pp. 1-18, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]