AI-Driven Crop Health Monitoring and Efficient Irrigation Management System

Abstract

This paper presents agricultural evolution using artificial intelligence and introduces automated crop monitoring and irrigation management system. Combining microcontroller, Internet of Things and deep learning, boosts the productivity and efficiency of farming in practice. A line follower robot captures plant images using a custom-trained YOLOv8 model locally deployed on a Raspberry Pi. The system monitors plant health by autonomously checking for plant diseases and nutrient deficiencies like Nitrogen, Phosphorus and Potassium (N, P, and K). A transfer learning approach was chosen with the use of VGG16 net for image analysis. The system also facilitates the user to give details about the crop that result in estimation of suitable amount of water required for healthy and reliable farming, thus empowering farmers with intuitive crop health monitoring and management for high yield.