A unified dataset for weed detection and segmentation in crops, optimized for YOLO models. It combines an original dataset of 9 common weed species from the León region with a portion of the Veridis dataset. The dataset was created by unileon-robotics and was last updated on Hugging Face in May 2026.
Use Cases
- Training object detection models for weed identification based on annotated images.
- Developing semantic segmentation models for precise weed localization in crop fields.
- Benchmarking YOLO-based models on agricultural computer vision tasks.
- Creating synthetic data pipelines by combining multiple weed image sources.
Strengths
- Combines data from two sources, likely increasing diversity.
- Specifically optimized for the popular YOLO model architecture.
- Focuses on 9 common weed species from a defined geographic region (León).
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count, file formats, and license are unknown, which may limit suitability assessment.
Provenance
- Source
- unileon-robotics on Hugging Face.
- Collection Method
- Combination of an original dataset and a portion of the Veridis dataset.
- Freshness
- Last updated 2026-05-05 08:46:34; freshness should be verified.
- Geography
- Weed species are common in the León region.