Exploring Machine Learning Techniques for Accurate Plant Leaf Disease Detection and Classification

Abstract

As a matter of fact, agriculture has been the very backbone of human civilization, not only being the strong engine of economic growth but also being the major source of food. Crop diseases, however, have become a grave threat to the health and productivity of crops, thus being a significant hindrance to agriculture and potential agricultural yields. There is now a need to detect and classify plant diseases in time and precisely with minimal infliction of further damage on crops. Methods that have been adopted by farmers to predict and classify diseases take long periods, and there is an element of error; hence, it is crucial to mechanize the disease forecasting process. Interfacing computerized techniques of image processing into agricultural fields has given tremendous hope as losses are reduced and productivity increased. Over the last few decades, scientists in their pursuit explored many methods for detection and classification of various types of plant diseases by looking at images of infected leaves or crops. This paper presents coverage of most recently developed technologies for detection and classification of various plant diseases with comparative study.

Plant phenotyping relevance

植物の病徴を画像から検出・分類する手法を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が中心です。

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.