Publication Type : Conference Paper
Publisher : Springer Nature Switzerland
Source : Communications in Computer and Information Science
Url : https://doi.org/10.1007/978-3-032-16038-6_2
Keywords : IoT, Smart irrigation, Crop recommendation, Soil detection, Disease prediction, ESP32, Firebase, Agriculture automation
Campus : Coimbatore
School : School of Engineering
Year : 2026
Abstract : This paper presents an integrated IoT-based system for automated irrigation and intelligent crop management, aimed at enhancing water use efficiency and agricultural productivity. The system employs an ESP32-based sensor node to monitor real-time soil moisture, NPK levels, temperature, and humidity. A Raspberry Pi acts as an edge gateway, processing data and storing it to a Firebase backend that synchronizes with a Flutter mobile application. The core of the system features three machine learning modules using a Random Forest classifier for crop recommendation with 98.48% accuracy, a MobileNetV2 model for image-based soil classification, and a CNN for okra leaf disease detection. It integrates crop management, soil classification, irrigation, and disease management in a unified application. The system's nutrient-sensing capabilities were validated against laboratory results from Tamil Nadu Agricultural University (TNAU), confirming its efficacy for real-time diagnostics. This work delivers a scalable, cost-effective, and user-friendly solution to support data-driven precision agriculture for small and marginal farmers.
Cite this Research Publication : C. P. Boopathy, S. Aravind, R. Adithya, S. SaiGanesh, S. M. Sharrath Mani, IoT-Based Automated Irrigation with Machine Learning for Crop Recommendation, Soil Classification, and Disease Detection, Communications in Computer and Information Science, Springer Nature Switzerland, 2026, https://doi.org/10.1007/978-3-032-16038-6_2
