Agri Sentinel AI: Revolutionizing Smart Agriculture with Real-Time Disease Detection and Advisory Systems

Abstract

Agriculture remains the backbone of many developing economies, yet it is increasingly threatened by plant diseases, insect infestations, and animal intrusions that lead to significant yield losses and economic instability for farmers. Conventional crop monitoring still depends largely on manual field visits and occasional expert supervision. This approach is labour intensive, slow, and unsuitable for continuous large-scale surveillance. In the early stages, symptoms of diseases such as rust, blight, or bacterial infections are often faint and easily overlooked, allowing problems to spread before any action is taken. To resolve these issues, this work introduces Agri Sentinel AI, an intelligent, real-time agricultural monitoring system that combines deep learning, computer vision, and Internet of Things (IoT) technologies. Low cost ESP32 camera modules are installed across the field to continuously capture live images of crops. These image streams are processed by YOLO-based object detection models capable of identifying plant diseases, pest activity, and animal intrusions under diverse environmental conditions. When a potential threat is spotted, the system produces labelled images highlighting detected areas and confidence levels, and links each detection to an integrated agricultural knowledge base. This knowledge base provides detailed disease descriptions, prevention strategies, and recommended treatments. The relevant information, along with the annotated image, is immediately delivered to farmers via Telegram alerts, enabling quick, informed responses. Field experiments conducted in real agricultural environments indicate that the system is accurate, scalable, and affordable, ensuring its suitability for smart farming and precision agriculture.

Introduction

The text presents Agri Sentinel AI, an AI- and IoT-based real-time crop monitoring system designed to help farmers detect plant diseases, insect infestations, and animal intrusions at an early stage. It highlights how traditional manual field inspections and existing mobile or survey-based tools are insufficient for continuous monitoring, especially under real-world farming conditions such as changing weather, lighting, large field sizes, and limited access to agricultural experts. These limitations often lead to delayed detection, increased pesticide use, higher costs, and significant crop losses.

To address these challenges, Agri Sentinel AI integrates ESP32 camera modules, YOLO-based deep learning object detection, and automated Telegram alerts into a single, affordable solution suitable for small and medium-scale farmers. The system continuously captures field images, analyzes them locally or at the edge for fast detection, and instantly notifies farmers with annotated images and practical advisory guidance drawn from an agricultural knowledge base.

Unlike prior research that focused mainly on laboratory-based disease classification or cloud-dependent systems, Agri Sentinel AI emphasizes real-time, outdoor deployment, low network dependency, and coverage of multiple threats including diseases, pests, and animals. Overall, the system aims to enable early intervention, reduce crop damage and pesticide overuse, and promote more precise, sustainable agricultural practices through continuous automated monitoring.

Conclusion

The proposed Agri Sentinel AI system successfully demonstrates the use of artificial intelligence and IoT for real-time plant disease and insect detection in agriculture. By integrating ESP32-CAM modules with a YOLO-based deep learning model, the system enables continuous monitoring of crop health with minimal human intervention. The implemented solution accurately detects diseases and pests, provides annotated visual feedback, and delivers timely alerts along with preventive and treatment recommendations through Telegram. Experimental results show that the system performs reliably under varying environmental conditions, making it suitable for practical agricultural deployment. Overall, the project proves that AI-driven monitoring can significantly assist farmers in early disease detection, reducing crop loss and improving productivity.

References

[1] E. A. Aldakheel, M. Zakariah and A. H. Alabdalall, “Detection and identification of plant leaf diseases using YOLOv4, statistical regression, and remote sensing,” Frontiers in Plant Science, 2024. [2] Performance Evaluation of YOLO Models in Plant Disease Detection, Journal of Intelligent Wireless Environments, 2024. [3] M. Shoaib, H. Alshammari, and A. Altamimi, “An advanced deep-learning model-based plant disease detection and classification,” Frontiers in Plant Science, 2023. [4] W. B. Demilie, T. T. Derseh, and M. B. Abebe, “Plant disease detection and classification techniques: a survey,” Journal of Big Data, 2024. [5] S. Kanakala and S. Ningappa, “Detection of leaf diseases in multi-crop leaves using LSTM and CNN,” arXiv preprint, 2025.

Copyright

Copyright © 2026 Manish M, Dr. Vidya E V, H D Dheemanth, Pradeep A, Ashwath Narayana Reddy N. 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.