Comparative performance of YOLOv8, YOLOv9, YOLOv10, YOLOv11 and Faster R-CNN models for detection of multiple weed species

• Developed annotated image database for cocklebur, dandelion, common waterhemp, common lambsquarters and Palmer amaranth to fine-tune YOLO and Faster R-CNN models • YOLOv11 was the fastest model with inference time of 13.5 ms, followed by YOLOv8 and YOLOv10 at 23 ms and 19.3 ms, respectively • YOLOv9 achieved highest detection accuracy with a mAP@0.5 of 0.935 • YOLO algorithms outperformed Faster R-CNN in speed and accuracy • YOLO models showed strong generalization from limited data and successfully distinguished between visually similar weed species, such as Palmer amaranth and common waterhemp. • Despite having fewer annotated images, dandelion and cocklebur had higher mAP values than Palmer amaranth and…