Comparative Analysis of YOLOv9, YOLOv10 and RT-DETR for Real-Time Weed Detection
It provides insights for selecting models to enhance agricultural productivity through precise weed management, but is incremental as it applies existing methods to a specific domain.
This paper compared YOLOv9, YOLOv10, and RT-DETR models for real-time weed detection in smart-spraying, evaluating mAP scores and inference times across various model sizes and image resolutions to highlight trade-offs between accuracy and speed.
This paper presents a comprehensive evaluation of state-of-the-art object detection models, including YOLOv9, YOLOv10, and RT-DETR, for the task of weed detection in smart-spraying applications focusing on three classes: Sugarbeet, Monocot, and Dicot. The performance of these models is compared based on mean Average Precision (mAP) scores and inference times on different GPU and CPU devices. We consider various model variations, such as nano, small, medium, large alongside different image resolutions (320px, 480px, 640px, 800px, 960px). The results highlight the trade-offs between inference time and detection accuracy, providing valuable insights for selecting the most suitable model for real-time weed detection. This study aims to guide the development of efficient and effective smart spraying systems, enhancing agricultural productivity through precise weed management.