Empowering early detection of plant diseases in agriculture using artificial intelligence

Abstract

Artificial Intelligence (AI) is revolutionizing plant disease diagnosis by providing transformative solutions to the challenges posed by agricultural diseases. AI-driven algorithms significantly reduce the time required for disease identification, enabling timely and precise control measures. These interventions help prevent the widespread proliferation of pathogens, mitigate crop losses and minimize economic damage. The integration of machine learning and deep learning particularly convolutional neural networks (CNNs) with computer vision systems enhances the precision, scalability and efficiency of disease monitoring. AI-powered tools offer real-time surveillance by capturing images of diseased leaves and generating data-driven insights, thereby facilitating targeted treatment applications while reducing resource wastage and environmental impact. Furthermore, the application of AI-powered mobile apps provides farmers with instant, field-level support to take preventive actions during the early stages of disease development. These technologies enable farmers to make informed, evidence-based decisions, optimize their agricultural practices and enhance crop yield and quality. Ultimately, AI plays a pivotal role in boosting agricultural productivity, ensuring food security and promoting both economic resilience and environmental sustainability. This review highlights recent advancements in machine learning algorithms, deep learning models especially CNNs and the role of mobile applications in early disease detection in agriculture.

Plant phenotyping relevance

植物病害の画像観察に基づくAI検出を主題とするレビューであり、植物の病徴・病害状態を推定する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 review article on AI for plant disease detection. It presents no authors' phenotype datasets, images, code, models, or supplements; all cited datasets/apps (e.g., PlantVillage, Ricexpert) are prior work or third-party tools, not paper-specific assets.

No evidence-backed public reproduction asset is currently recorded.