PLANT LEAF DISEASE DETECTION

Abstract

ABSTRACT- Agriculture remains a fundamental pillar of many national economies, making the protection of crops from disease a top priority. Pathogens such as bacteria, fungi, and viruses can significantly reduce crop productivity, underscoring the need for timely and accurate disease detection. Recent innovations in computer vision and artificial intelligence have introduced powerful tools for recognizing plant diseases through image analysis, particularly using leaf imagery. This paper investigates the application of machine learning, deep learning, and few-shot learning models in automating disease identification to assist farmers in making informed, prompt decisions. By examining the use of advanced models—including convolutional networks and vision transformers—alongside imaging technologies like hyperspectral cameras, this study highlights both the technological advancements and their potential impact in the field. Furthermore, it touches on molecular-level diagnostic techniques aimed at minimizing the threat of pathogens. The review offers a thorough overview of current progress and identifies key opportunities for future research, with the goal of translating laboratory breakthroughs into practical solutions for sustainable agriculture. INDEX TERMS: Plant disease, deep learning, machine learning, shot learning, computer vision, folding networks (CNNS), vision trans, hyperspectral imaging, molecular diagnostics, sustainable agriculture detection.

Plant phenotyping relevance

植物葉画像から病害状態を推定する画像解析・機械学習手法を中心に扱うレビューであり、植物病害フェノタイピング手法のレビューに該当する。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.