Advances in Molecular, Digital, and Remote Sensing Technologies for Early Crop Disease Detection: A Comprehensive Review

Abstract

Crop diseases caused by diverse pathogens, including fungi, bacteria, and viruses, lead to result in global yield losses of nearly 20–40% each year, posing a major significant threat to food security and agricultural sustainability. Traditional detection methods, relying on visual inspection and routine laboratory assays, are often slow, labour-intensive, and prone to inaccuracies, resulting in delayed disease management. By enabling quick, precise, and scalable detection systems, recent advancements in molecular biology, digital technologies, and remote sensing have completely transformed the field of crop disease diagnostics. Molecular techniques such as real-time PCR, loop-mediated isothermal amplification (LAMP), and CRISPR-based assays (e.g., SHERLOCK) offer high sensitivity and specificity, allowing early on-site pathogen identification. Digital technologies driven by artificial intelligence, including smartphone-based diagnostic tools and convolutional neural networks (CNNs), now achieve over 95% accuracy in image-based disease recognition, making advanced diagnostics more accessible to farmers. Remote sensing approaches particularly drone-assisted hyperspectral and multispectral imaging facilitate non-invasive, large-scale monitoring and early detection of disease outbreaks across agricultural landscapes. Additionally, metagenomics and next-generation sequencing (NGS) enable the discovery of novel pathogens and support resistance-breeding programs through comprehensive genomic insights. Collectively, these innovative technologies enhance the speed, precision, and cost-effectiveness of crop disease detection, potentially reducing yield losses by up to 30% and promoting sustainable agriculture. This review highlights the principles, recent advancements, advantages, limitations, and prospects of integrating molecular, digital, and remote sensing tools to strengthen global crop health management systems.

Code and data availability

This is a review article with no original phenotyping measurements, datasets, images, code, or models; no public paper-specific assets are mentioned.

No evidence-backed public reproduction asset is currently recorded.