Advancing Plant Diseases Detection with Pre-trained YOLO Models
Boudjemaa Boudaa, Kamel Abada, Walid Aymen Aichouche, Ahmed Nabil Belakermi
Abstract
The current paper explores the close interaction between technological advancements and the field of agriculture through a project focused on plant disease detection using Deep Learning techniques. It provides an in-depth analysis of the plant disease phenomena and its negative impact on crops, while also highlighting the effectiveness of using pre-trained models of YOLO (You Only Look Once) in the agricultural domain for plant disease detection. A comparative study is conducted on different important versions of YOLO (v5, v8 and v9) on a real-world dataset. This study was achieved by recommending the application of deep learning models mainly YOLOv9 to enhance detection accuracy and improve agricultural productivity overall.