Yolo and Its Evolved Versions: A Survey on Feature Enhancements for Improved Plant Disease Detection

Abstract

The agricultural sector has increasingly recognized the significance of integrating computer vision and machine learning in recent years. Computer vision (CV) technology has transformed farming through the development of autonomous, scalable sensor systems . These technologies, which use remote cameras and advanced CV algorithms, have several uses, from lowering production costs through intelligent automation to improving overall efficiency. In agriculture, one of the most significant challenges is accurately detecting plant leaf diseases, which can significantly affect crop quality and yield. One significant advancement in this area is the You Only Look Once (YOLO) framework, a state-of-the-art object identification method that formulates detection as a single regression problem. YOLO can recognize many disease types in a single image with speed and accuracy. This study presents a thorough analysis of plant disease detection methods based on multiple YOLO versions. It explains and evaluates enhancements made to the original YOLO design, summarizes the findings of earlier studies where constructively looks at performance metrics, and discusses potential directions for future development.

Plant phenotyping relevance

植物葉の病害状態を画像から検出するYOLO手法を比較・評価するレビューであり、病害表現型の取得・推定手法が中心である。

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.