AI Doctor for Plants Using CNN and Raspberry Pi

Abstract

In recent years, agriculture has faced several challenges, especially regarding plant health monitoring and disease detection. Traditional manual inspection methods are time-consuming, labor-intensive, and often inaccurate. This paper proposes an intelligent solution an “AI Doctor for Plants” that utilizes Convolutional Neural Networks (CNN) deployed on a Raspberry Pi system to detect and classify plant diseases in real time. The model captures leaf images using a camera module, processes them locally through a trained CNN model, and displays the identified disease along with suggested preventive measures. This project minimizes the dependency on internet connectivity by using TensorFlow Lite, enabling offline disease detection. The proposed system helps farmers and gardeners to identify diseases early, improve crop yield, and reduce chemical usage

Introduction

Agriculture remains essential to food security, especially in developing countries where early detection of plant diseases is critical to preventing major yield and economic losses. Traditional diagnosis depends on expert evaluation, which is not always accessible to farmers in remote regions. To address this gap, the AI Doctor for Plants system uses Artificial Intelligence, particularly Convolutional Neural Networks (CNNs), combined with Raspberry Pi hardware to automate plant disease detection through leaf image analysis.

The methodology involves capturing leaf images with a Raspberry Pi camera, preprocessing them, and classifying them using a CNN model trained on datasets such as PlantVillage. The model is optimized using TensorFlow Lite to enable fast, offline inference on the Raspberry Pi. After classification, the system displays the detected disease and provides recommended treatments and preventive measures.

Existing research shows high accuracy in plant disease classification using CNNs, with studies reporting up to 99% accuracy. The proposed system builds on these advancements by integrating IoT capabilities for remote monitoring, environmental data collection via sensors, and optional cloud connectivity for long-term storage and analysis.

The prototype is expected to achieve 92–96% accuracy with real-time processing under two seconds per image. It offers an affordable and accessible solution for farmers, supporting both early disease detection and preventive agriculture. Future enhancements may include integration with mobile apps, cloud platforms, IoT-based environmental monitoring, solar-powered units for field deployment, and expansion to detect nutrient deficiencies and pest infestations.

Conclusion

The AI Doctor for Plants presents a promising approach to integrating deep learning into agricultural diagnostics. It enables cost-effective, accurate, and offline plant disease detection using CNNs on Raspberry Pi. The project empowers farmers with technology-driven insights, helping to improve productivity and sustainability. Future work can involve expanding the model to include more crop species, integrating soil and weather data, and developing a mobile application interface for broader accessibility.

References

[1] Mohanty, S. P.; Hughes, D. P.; Salathé, M. “Using Deep Learning for Image-Based Plant Disease Detection.” Frontiers in Plant Science, vol. 7, 2016. Available at: https://doi.org/10.3389/fpls.2016. [2] Upadhyay, A.; Chandel, N. S.; Singh, K. P.; et al. “Deep learning and computer vision in plant disease detection: a comprehensive review of techniques, models, and trends in precision agriculture.” Artificial Intelligence Review, vol. 58, article 92, 2025. Available: https://doi.org/10.1007/s10462-024-11100-x [3] Katafuchi, R.; Tokunaga, T. “Image-based Plant Disease Diagnosis with Unsupervised Anomaly Detection Based on Reconstructability of Colors.” arXiv preprint, 2020. Available at: https://arxiv.org/abs/2011.14306 [4] Cap, Q. H.; Uga, H.; Kagiwada, S.; Iyatomi, H. “LeafGAN: An Effective Data Augmentation Method for Practical Plant Disease Diagnosis.” arXiv preprint, 2020. Available at: https://arxiv.org/abs/2002.10100 [5] Singh, P.; Khanna, P.; Sheorey, T.; Ojha, A. “Explainable Vision Transformer Enabled Convolutional Neural Network for Plant Disease Identification: PlantXViT.” arXiv preprint, 2022. Available at: https://arxiv.org/abs/2207.07919

Copyright

Copyright © 2025 Ms. Bhakti Satish Powar, Ms. Sakshi Mahesh Lohar, Ms. Samiksha Arun Powar, Ms. Priyanka Pramod Sartape, Mr. Shital Narayan Gavade. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.