Analysis of IoT-Collected Radish Growth Data Using Deep Neural Networks

Managing crop growth conditions with an eye toward sustainability helps optimize resource use and reduce environmental impact. The aim of this study was to apply genetic algorithm-based hyperparameter optimization (GA-HPO) to deep neural networks (DNNs) to classify an IoT-collected dataset on radish growth conditions, with the goal of identifying optimal growth parameters. By using physiological and morphological data from various radish species and treatment conditions, this research explores the effectiveness of GA in fine-tuning DNN hyperparameters to achieve high classification accuracy. The GA-HPO approach improved model performance to a classification accuracy of 92%, showcasing its adaptability and potential as a tool for precision agriculture. Through the identification of optimal growing conditions, this study supports sustainable practices in agriculture by facilitating resource-efficient and environmentally friendly crop management.