IOT-Based Smart Plant Monitoring System

Smart agriculture systems enhance crop productivity through automation, real-time monitoring, and intelligent decision-making. This paper presents the design and implementation of an intelligent plant identification and leaf disease detection system integrated with automated irrigation and chemical spraying mechanisms using a Raspberry Pi-based embedded platform. A lightweight YOLO-based deep learning model (YOLOv8 Nano) is trained offline using a labeled dataset of healthy and diseased plant leaves. The training process is performed on a high-performance computing environment to achieve high detection accuracy. The model is optimized for edge deployment and subsequently deployed on the Raspberry Pi for real-time inference using a camera module. The system detects plant leaves, classifies disease types, and identifies infected regions through bounding box detection. When the detected disease confidence exceeds a predefined threshold, an automated chemical spraying mechanism is activated to prevent further spread. Environmental parameters such as temperature and humidity are monitored using a DHT22 sensor, while soil moisture levels are measured using a soil moisture sensor. Based on real-time soil moisture readings, irrigation is automatically controlled using a DC pump operated through a relay module to ensure efficient water utilization. All sensor data, detection results, actuator status, and system parameters are displayed on a web-based dashboard for remote monitoring and control. The proposed system reduces manual intervention, enables early disease diagnosis, optimizes water management, and supports precision agriculture practices. Experimental results demonstrate reliable real-time performance and effective disease detection on a resource-constrained edge device, making the system suitable for smart farming applications.