hotosm/fAIr-guru-td-trees

fAIr Open Mapping Gurus: trees

Tree training data from the Open Mapping Gurus programme, built from a Freetown, Sierra Leone project that mapped trees to generate AI training data. Drone imagery from OpenAerialMap is paired with OpenStreetMap trees mapped and validated in HOT Tasking Manager.

At a glance

Tiles	48,490
Splits	train 38,705, validation 4,158, test 5,627
Split method	10 contiguous regions per project, 1 region to test, 1 region to validation; tiles at a region edge border tiles of another split; a feature crossing an edge has parts in both splits
Tiles with trees	35,282
Tiles without trees	13,208
Trees	120,434
Ground pixel size	0.15 m (zoom 20)
Area covered	68.85 sq km
Countries	Sierra Leone
Version	0.1.1, built 2026-10-20

Sample tiles

Each tile is shown beside its labels, coloured by class, with the main class and its count above.

Classes

Labels are OpenStreetMap features tagged natural=tree or natural=tree_group, and a label's class is its natural value. The wiki describes how each tag is mapped: Key:natural, Tag:natural=tree, Tag:natural=tree_group.

Class	Labels	Share
tree	95,992	79.7%
tree_group	24,442	20.3%

Source projects

Project	Country	Tiles	Trees	Ground pixel size
50661: FREETOWN TREE and Solar Panel MAPPING	Sierra Leone	48,490	120,434	0.15 m

How it was labelled

The labels are OpenStreetMap features mapped in HOT Tasking Manager projects (wiki). Tasking Manager divides a project area into small tasks, so many volunteers can map it together following the project instructions. A mapper traces the features in a task, and a validator then checks the task and marks it validated. Coordinated mapping of this kind falls under the OpenStreetMap Foundation's Organised Editing Guidelines.

The imagery comes from OpenAerialMap (wiki), where people and organisations share openly licensed satellite, aerial and drone imagery through the Open Imagery Network. Each project sets this OpenAerialMap imagery for its mappers, and the tiles come from it.

The project lists OTHER as its mapping type and instructed mappers to map isolated trees as points tagged natural=tree, with an estimated crown diameter in diameter_crown, and dense or overlapping trees as polygons tagged natural=tree_group. Tiles come only from task squares marked VALIDATED. Labels are every OpenStreetMap feature with natural=tree or natural=tree_group intersecting a tile.

Labels are published as mapped, points and polygons in label_geojson, and the mask column is empty, so users can build masks to suit their model, for example by buffering points with diameter_crown.

Splits

The splits are spatial. Each project's tiles are grouped into contiguous regions, and each region goes whole to train, validation or test, as listed under Split method above. For model evaluation we recommend making your own split, as described in Make your own split.

Load

import json

from datasets import load_dataset
from rasterio.features import rasterize
from rasterio.transform import from_bounds

dataset = load_dataset("hotosm/fAIr-guru-td-trees", split="train")
labelled = dataset.filter(lambda count: count > 0, input_columns="num_features")
row = labelled[0]
image = row["image"]
labels = json.loads(row["label_geojson"])

bounds = (row["bbox_west"], row["bbox_south"], row["bbox_east"], row["bbox_north"])
transform = from_bounds(*bounds, image.width, image.height)
shapes = [feature["geometry"] for feature in labels["features"]]
mask = rasterize(shapes, out_shape=(image.height, image.width), transform=transform)

image is the 256x256 tile as served by OpenAerialMap. The bbox_* columns hold its bounds, which place label_geojson on its pixels.

Column	Description
image	RGB tile
mask	Binary raster of the labels, 0 background and 255 feature, in datasets built with masks
label_geojson	Labels in the tile with their OpenStreetMap tags, EPSG:4326
num_features, class_counts	Label count, and the count per class as JSON
tile_id, tile_x, tile_y, tile_z	XYZ tile identity
bbox_west, bbox_south, bbox_east, bbox_north	Tile bounds, EPSG:4326
project_id, project_name, project_url, country, organisation	Source Tasking Manager project
imagery_url, feature_type	OpenAerialMap tile URL, and the OpenStreetMap key of the classes

Browse the STAC catalogue

Each project is one item holding a Cloud Optimized GeoTIFF of its tiles, the labels and the tile footprints with their split. Open in STAC Browser to see the labels over the imagery, or follow the QGIS guide to open them in QGIS.

Known limitations

• diameter_crown is a mapper estimate. Of the trees mapped in the project area, about 57% give it with a unit in varying formats (3m, 2.412m, 3.234 m, 3,134 m), 38% as a bare number and 4% leave it out, and many values repeat exactly, so it gives an approximate crown size.
• Some mappers marked shrubs and small plants as trees; the project instructions ask for trees only.
• A tree group crossing a tile edge appears in each tile it touches, so label counts are per tile.

Sources and licensing

Component	Source	License
Imagery	OpenAerialMap	CC-BY 4.0, attribute contributors of Open Imagery Network
Labels	OpenStreetMap	ODbL 1.0, attribute OpenStreetMap contributors

Built with fAIr-ds-builder.