globusjournal

This study presents an Internet of Things (IoT) approach to hydroponic cultivation, integrating sensor technology with machine learning algorithms to enhance crop management and harvest timing. Utilizing a Node-RED architecture, the system processes data from sensors measuring pH, total dissolved solids (TDS), temperature, and humidity, enabling real-time monitoring and analysis of growing conditions. The research applies machine learning models to assess plant health and determine harvest readiness. The model shows varying effectiveness across disease categories, with performance metrics highlighting strengths and areas for improvement. Notably, the binary classification model for harvest readiness demonstrates balanced performance, potentially improving harvest timing decisions. By combining IoT sensor technology with predictive analytics, this system aims to improve efficiency and reduce resource consumption in controlled environment agriculture. The findings contribute to the field of smart agriculture, suggesting directions for further research in precision agriculture and sustainable food production.

Keywords: IoT, Hydroponics, Disease, Harvestability, Lettuce.