A Review of Image Processing and Machine Learning for Plant Leaf Disease Identification

Abstract

One of the main factors affecting crop output reduction globally is plant disease and crop losses must be avoided through early detection of these diseases.Automating the identification of plant diseases has been demonstrated to be highly promising when using image processing and machine learning approaches.In the present work, fusion techniques were utilised to combine data from several sources to increase the reliability and accuracy of identifying plant leaf diseases.Fusion approaches combine information from several sources, such as various pictures or feature kinds, to produce a more complete view of the plant leaf and its illness.Constructing a trustworthy and accurate system that can automatically recognise the symptoms of diseases in tomato leaves utilising several sources of data including pictures, spectral reflectance, and environmental parameters is the major objective of tomato leaf disease detection using machine learning.This approach can assist farmers and agricultural professionals in identifying the disease early, stopping it from spreading and acting quickly to reduce crop losses.

Plant phenotyping relevance

植物葉の病徴を画像処理・機械学習で識別する手法のレビューであり、植物病害状態の画像ベース表現型計測が中心です。

Code and data availability

This is a review paper with no public datasets, images, code, models, or supplements of its own; cited works are prior studies, not paper-specific assets.

No evidence-backed public reproduction asset is currently recorded.