--- license: mit language: - en - pt library_name: ultralytics pipeline_tag: image-segmentation base_model: Ultralytics/YOLOv8 datasets: - duartepcruz/acacia_dataset tags: - yolov8-seg - instance-segmentation - remote-sensing - satellite-imagery - invasive-species - forest-monitoring model-index: - name: forest-guardian-yolov8-seg results: - task: type: image-segmentation dataset: name: acacia_dataset (validation split) type: duartepcruz/acacia_dataset metrics: - type: mAP@50 (box) value: 0.56 - type: mAP@50 (mask) value: 0.24 --- # Forest Guardian: YOLOv8n-seg for invasive acacia Segments invasive acacia stands in top-down satellite imagery of Portugal. 🥇 1st place overall, JunctionX Lisbon 2025. **[▶ Try it in your browser](https://duartepcruz.com/work/invasive-species-mapping/)** · [Code](https://github.com/DuarteCruz6/JunctionX-Lisbon2025-Hackathon) · [Dataset](https://huggingface.co/datasets/duartepcruz/acacia_dataset) ## Model | | | |---|---| | Architecture | YOLOv8n-seg (3.4M params) | | Training | Fine-tuned 15 epochs at 640 px | | Formats | PyTorch `.pt`; ONNX export (13 MB) used in the browser demo | ## Results (validation split) | Metric | Value | |---|---| | Box mAP@50 | 0.56 | | Mask mAP@50 | 0.24 | | Acacia tiles flagged | 22 / 32 | | Negative tiles flagged | 0 / 16 | ## Training data Satellite tiles (Esri World Imagery) cut around acacia stands mapped in Portugal's COS2018 land-use map (class 5.1.1.6), plus tiles of other land as negatives. Masks are COS2018 polygons rasterised to 640×640. See the [dataset card](https://huggingface.co/datasets/duartepcruz/acacia_dataset). ## Usage ```python from huggingface_hub import hf_hub_download from ultralytics import YOLO weights = hf_hub_download("duartepcruz/forest-guardian-yolov8-seg", "best.pt") model = YOLO(weights) results = model.predict("tile.jpg", imgsz=640, conf=0.25) results[0].show() ``` ## Intended use This is a first pass that points foresters at places worth checking. It is tuned to stay quiet when unsure, so **an empty result means "nothing obvious", not "no acacia"**. ## Limitations - **Recall:** it misses about a third of acacia stands, especially in yellow bloom (most training tiles show non-flowering trees). Lowering `conf` to about 0.15 helps. - **Coarse masks:** the labels are map polygons at landscape scale, not tree crowns, so the boxes are more reliable than the mask outlines. - **Domain:** it was trained on a single satellite basemap at a similar zoom. Drone imagery, other seasons and other resolutions are out of distribution. - **Labels:** COS2018 is a land-use map, not field-verified ground truth.