Advances in UAV and AI Applications for Crop Disease Monitoring

Abstract

Plant diseases are estimated to cause a reduction of 20–40% in worldwide crop yields leading to over USD 220 billion in lost productivity and food security each year. Detecting these diseases early is essential for sustainable management, however traditional methods like scouting and laboratory diagnostics are often slow and impractical for large-scale or pre-symptomatic monitoring. This review looks at recent developments in using Unmanned Aerial Vehicles (UAVs) combined with Artificial Intelligence (AI) for overseeing crop health. It compares different sensor types such as RGB, multispectral, hyperspectral, thermal and LiDAR and explains the process from data collection to AI-driven classification. A particular focus is on machine learning (ML) and deep learning (DL) including Convolutional Neural Network (CNN) architectures, which have achieved 90–98% accuracy in identifying diseases in crops like wheat, potatoes, citrus and grapevines. The review further explores exciting new directions like data fusion, edge computing and autonomous scouting, pointing towards a future of more proactive, scalable and precise disease management. Keywords: Climate change, hi-tech agriculture, remote sensing, pre-symptomatic, autonomous scouting, machine learning.

Plant phenotyping relevance

UAVセンサーとAIによる作物病害状態の観測・分類手法を中心に比較・レビューしており、植物フェノタイピング手法のレビューに該当する。

Code and data availability

This is a narrative review of UAV/AI crop disease monitoring with no original phenotyping measurements or analysis. The authors explicitly state no new data were generated or analyzed, and no paper-specific datasets, images, code, or models are deposited; all cited works are prior publications, not assets of this paper

No evidence-backed public reproduction asset is currently recorded.