Smart Greenhouse Management: Harnessing Artificial Intelligence for Sustainable Farming

Abstract

Abstract Greenhouse farming plays a vital role in enhancing agricultural productivity, yet it often suffers from inefficient resource management and delayed disease detection. This paper presents a novel solar-powered Smart Greenhouse Management System (SGHMS) that integrates IoT-based environmental monitoring, machine learning for real-time disease detection, and a Raspberry Pi-controlled autonomous sprayer into a unified platform. Unlike existing systems, our approach combines a CNN-based plant health classifier deployed locally on Raspberry Pi with an energy-efficient solar power source to ensure reliable off-grid operation. A user-friendly web and mobile application enables real-time monitoring, alert generation, and remote control of environmental parameters and spraying actions. The system was deployed in a real greenhouse for 30 days and demonstrated a 92% disease detection accuracy while significantly reducing water and energy consumption. This integrated solution offers a scalable and cost-effective approach to sustainable precision agriculture, particularly in resource-constrained regions.

Plant phenotyping relevance

CNNによる植物健康・病害検出を中核機能として実 greenhouse で展開し、検出精度も評価しているため、植物状態の画像ベース表現型計測を含む実質的なプラットフォーム研究である。

Code and data availability

The paper describes a solar-powered smart greenhouse system with a CNN disease detector, but provides no public dataset, images, code, or model. Code availability is 'Not applicable' and data are 'available upon request' with no URL, so no qualifying public asset exists.

No evidence-backed public reproduction asset is currently recorded.