A Multi-Dataset Benchmark of YOLO-Based Weed Detection in Precision Agriculture

2026-09-27

AI Summary

This study addresses the limited cross-domain robustness of existing weed detection models, which typically rely on single-dataset evaluation. We construct the first multi-source YOLO benchmark encompassing seven agricultural scenarios to systematically evaluate model performance in accuracy, efficiency, and cross-domain generalization. Methodologically, we employ YOLOv2 through YOLOv6 variants and design a multi-source joint training protocol with cross-domain testing to quantify the impact of domain shift. Experimental results demonstrate that while YOLOv6s effectively balances accuracy and inference speed, its cross-domain mean Average Precision (mAP) degrades significantly, underscoring the urgent need for domain adaptation techniques. This work provides the first systematic cross-domain evaluation framework and empirical evidence for object detection in precision agriculture.

Abstract

Weed detection is an important component of precision agriculture, enabling site-specific weed management and reducing unnecessary herbicide use. Although deep learning methods have achieved strong results for crop and weed detection, many studies rely on single-dataset evaluation, making it difficult to assess robustness across different agricultural domains. This paper presents a multi-dataset benchmark of deep object detectors for weed detection in precision agriculture, with a focused evaluation of YOLO26 models. We evaluate nano, small, and medium variants on seven public weed-detection datasets covering different crops, weed species, field conditions, acquisition setups, and annotation protocols. The models are compared in terms of detection accuracy, model complexity, inference latency, FPS, and model size. In addition to in-dataset evaluation, we investigate cross-domain generalization using a unified one-class weed setup and evaluate multi-source training using the combined training subsets from all datasets. The results show that YOLO26 achieves strong in-dataset performance, with YOLO26m obtaining the highest average accuracy and YOLO26s providing the best practical accuracy-efficiency trade-off. However, cross-domain performance decreases substantially, with YOLO26s dropping from an average in-domain mAP$_{50:95}$ of 0.603 to 0.148 in the off-domain setting. Multi-source training improves performance on several datasets, but does not fully eliminate domain shift. Overall, the benchmark highlights the importance of dataset diversity, domain similarity, and target-domain adaptation for robust weed detection in real-world precision agriculture applications.

Problem

• weed detection
• precision agriculture
• cross-domain generalization
• multi-dataset benchmark
• domain shift

Innovation

• Multi-dataset benchmark
• Weed detection
• YOLO26
• Cross-domain generalization
• Multi-source training