Spot spraying represents an efficient and sustainable method for reducing herbicide use in agriculture. Reliable differentiation between crops and weeds, including species-level classification, is essential for real-time application. This study compares state-of-the-art object detection models-YOLOv8, YOLOv9, YOLOv10, and RT-DETR-using 5611 images from 16 plant species. Two datasets were created, dataset 1 with training all 16 species individually and dataset 2 with grouping weeds into monocotyledonous weeds, dicotyledonous weeds, and three chosen crops. Results indicate that all models perform similarly, but YOLOv9s and YOLOv9e, exhibit strong recall (66.58 % and 72.36 %) and mAP50 (73.52 % and 79.86 %), and mAP50-95 (43.82 % and 47.00 %) in dataset 2. RT-DETR-l, excels in precision reaching 82.44 % (dataset 1) and 81.46 % (dataset 2) making it ideal for minimizing false positives. In dataset 2, YOLOv9c attains a precision of 84.76% for dicots and 78.22% recall for Zea mays L.. Inference times highlight smaller YOLO models (YOLOv8n, YOLOv9t, and YOLOv10n) as the fastest, reaching 7.64 ms (dataset 1) on an NVIDIA GeForce RTX 4090 GPU, with CPU inference times increasing significantly. These findings emphasize the trade-off between model size, accuracy, and hardware suitability for real-time agricultural applications.

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.

Spot spraying can significantly reduce herbicide use while maintaining equal weed control efficacy as a broadcast application of herbicides. Several online spot-spraying systems have been developed, with sensors mounted on the sprayer or by recording the RTK-GNSS position of each crop seed. In this study, spot spraying was realized offline based on georeferenced unmanned aerial vehicle (UAV) images with high spatial resolution. Studies were conducted in four maize fields in Southwestern Germany in 2023. A randomized complete block design was used with seven treatments containing broadcast and spot applications of pre-emergence and post-emergence herbicides. Post-emergence herbicides were applied at 2–4-leaf and at 6–8-leaf stages of maize. Weed and crop density, weed control efficacy (WCE), crop losses, accuracy of weed classification in UAV images, herbicide savings and maize yield were measured and analyzed. On average, 94% of all weed plants were correctly identified in the UAV images with the automatic classifier. Spot-spraying achieved up to 86% WCE, which was equal to the broadcast herbicide treatment. Early spot spraying saved 47% of herbicides compared to the broadcast herbicide application. Maize yields in the spot-spraying plots were equal to the broadcast herbicide application plots. This study demonstrates that spot-spraying based on UAV weed maps is feasible and provides a significant reduction in herbicide use.

Effective weed management is crucial in the critical period of sugar beet production, but often lacks sustainability and environmental protection. Recent advancements in sensor-based weed control systems have rendered the latter a realistic prospect, which demands detailed analyses, especially under suboptimal field conditions. The present study analysed six robotic-assisted weed control systems (RAWS) in three experiments on sugar beets in 2024, conducted under dry soil and high weed pressure. The experiments included sensor-based inter-row and intra-row hoeing, spot-and band-spraying and were compared to a broadcast herbicide treatment and an untreated control. Weed control efficacy (WCE) in the intra-and inter-row areas, as well as weed species composition and crop plant damage, were assessed after treatment. The data show that intra-row WCE of two hoeing robots (Farming GT® and Robovator®) equipped with

Spot spraying represents an efficient and sustainable method for reducing the amount of pesticides, particularly herbicides, used in agricultural fields. To achieve this, it is of utmost importance to reliably differentiate between crops and weeds, and even between individual weed species in situ and under real-time conditions. To assess suitability for real-time application, different object detection models that are currently state-of-the-art are compared. All available models of YOLOv8, YOLOv9, YOLOv10, and RT-DETR are trained and evaluated with images from a real field situation. The images are separated into two distinct datasets: In the initial data set, each species of plants is trained individually; in the subsequent dataset, a distinction is made between monocotyledonous weeds, dicotyledonous weeds, and three chosen crops. The results demonstrate that while all models perform equally well in the metrics evaluated, the YOLOv9 models, particularly the YOLOv9s and YOLOv9e, stand out in terms of their strong recall scores (66.58 % and 72.36 %), as well as mAP50 (73.52 % and 79.86 %), and mAP50-95 (43.82 % and 47.00 %) in dataset 2. However, the RT-DETR models, especially RT-DETR-l, excel in precision with reaching 82.44 \% on dataset 1 and 81.46 % in dataset 2, making them particularly suitable for scenarios where minimizing false positives is critical. In particular, the smallest variants of the YOLO models (YOLOv8n, YOLOv9t, and YOLOv10n) achieve substantially faster inference times down to 7.58 ms for dataset 2 on the NVIDIA GeForce RTX 4090 GPU for analyzing one frame, while maintaining competitive accuracy, highlighting their potential for deployment in resource-constrained embedded computing devices as typically used in productive setups.