AI in Agriculture: Enhancing Smart Farming

Abstract

Agriculture plays a critical role in ensuring food secu- rity and
sustaining economic growth, yet traditional farming practices often
rely on manual observation and experience- based decision-making,
which can result in delayed disease detection, inaccurate yield
estimation, and inefficient resource utilization. To address these
challenges, this paper presents a cloud-based Artificial Intelligencedriven Smart Agricul- ture system that integrates Internet of Things
(IoT) sensors, machine learning algorithms, and deep learning
techniques within a scalable cloud environment. The proposed system
continuously collects real-time environmental parameters such as
temperature, humidity, and soil moisture using IoT de- vices and
processes the data through predictive models for crop suitability
analysis and yield prediction. In addition, a MobileNetV2-based
convolutional neural network is imple- mented for automated tomato
leaf disease classification with confidence-based detection. The entire
framework is deployed using Amazon Web Services (AWS) to ensure
secure data storage, real-time processing, and scalable architecture. By
combining IoT monitoring, predictive analytics, and image- based
disease detection into a unified platform, the proposed
solution enhances early disease identification, improves crop planning
accuracy, and supports precision farming practices. Experimental
evaluation demonstrates reliable prediction per- formance,
highlighting the effectiveness ofthe integrated smart agriculture system
for sustainable farming applications.