Proof Of Concept: Integrating Deep Learning And IOT In Oyster Mushroom Farming

Abstract

This proof-of-concept study explores the integration of deep learning and Internet of Things (IoT) technologies in oyster mushroom (Pleurotus ostreatus) farming to overcome challenges of manual monitoring, inconsistent yields, and resource inefficiency. The objective was to design and evaluate an intelligent system for automated environmental control and harvest prediction. Guided by the design thinking framework, the study progressed from problem identification to prototype development and testing. A solar-powered mushroom house (4 × 6 ft) was equipped with IoT sensors for temperature, humidity, air quality, and light, integrated with actuators for misting, ventilation, and lighting. Convolutional neural network models, trained using collected and historical datasets, enabled accurate harvest readiness prediction. Experimental results showed prediction accuracy of 85–92% and effective environmental regulation with reduced human intervention. Despite challenges related to sensor calibration, dataset size, and scalability, the findings validate the feasibility of IoT–deep learning integration. The novelty of this work lies in combining real-time sensing with predictive analytics, offering a sustainable and scalable pathway toward intelligent mushroom farming for smallholder agriculture.