Train a Building Footprints Detection Model for the USA¶

Install package¶

To use the geoai-py package, ensure it is installed in your environment. Uncomment the command below if needed.

In [ ]:

# %pip install geoai-py

# %pip install geoai-py

Import libraries¶

In [ ]:

import geoai

import geoai

Download sample data¶

In [ ]:

train_raster_url = (
    "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_rgb_train.tif"
)
train_vector_url = "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_train_buildings.geojson"
test_raster_url = (
    "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_test.tif"
)

train_raster_url = ( "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_rgb_train.tif" ) train_vector_url = "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_train_buildings.geojson" test_raster_url = ( "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_test.tif" )

In [ ]:

train_raster_path = geoai.download_file(train_raster_url)
train_vector_path = geoai.download_file(train_vector_url)
test_raster_path = geoai.download_file(test_raster_url)

train_raster_path = geoai.download_file(train_raster_url) train_vector_path = geoai.download_file(train_vector_url) test_raster_path = geoai.download_file(test_raster_url)

Visualize sample data¶

In [ ]:

geoai.view_vector_interactive(train_vector_path, tiles=train_raster_url)

geoai.view_vector_interactive(train_vector_path, tiles=train_raster_url)

In [ ]:

geoai.view_raster(test_raster_url)

geoai.view_raster(test_raster_url)

Create training data¶

In [ ]:

out_folder = "output"
tiles = geoai.export_geotiff_tiles(
    in_raster=train_raster_path,
    out_folder=out_folder,
    in_class_data=train_vector_path,
    tile_size=512,
    stride=256,
    buffer_radius=0,
)

out_folder = "output" tiles = geoai.export_geotiff_tiles( in_raster=train_raster_path, out_folder=out_folder, in_class_data=train_vector_path, tile_size=512, stride=256, buffer_radius=0, )

Train object detection model¶

In [ ]:

geoai.train_MaskRCNN_model(
    images_dir=f"{out_folder}/images",
    labels_dir=f"{out_folder}/labels",
    output_dir=f"{out_folder}/models",
    num_channels=3,
    pretrained=True,
    batch_size=4,
    num_epochs=100,
    learning_rate=0.005,
    val_split=0.2,
)

geoai.train_MaskRCNN_model( images_dir=f"{out_folder}/images", labels_dir=f"{out_folder}/labels", output_dir=f"{out_folder}/models", num_channels=3, pretrained=True, batch_size=4, num_epochs=100, learning_rate=0.005, val_split=0.2, )

Run inference¶

In [ ]:

masks_path = "naip_test_prediction.tif"
model_path = f"{out_folder}/models/building_footprints_usa.pth"

masks_path = "naip_test_prediction.tif" model_path = f"{out_folder}/models/building_footprints_usa.pth"

In [ ]:

geoai.object_detection(
    test_raster_path,
    masks_path,
    model_path,
    window_size=512,
    overlap=256,
    confidence_threshold=0.5,
    batch_size=4,
    num_channels=3,
)

geoai.object_detection( test_raster_path, masks_path, model_path, window_size=512, overlap=256, confidence_threshold=0.5, batch_size=4, num_channels=3, )

Vectorize masks¶

In [ ]:

output_path = "naip_test_prediction.geojson"
gdf = geoai.orthogonalize(masks_path, output_path, epsilon=2)

output_path = "naip_test_prediction.geojson" gdf = geoai.orthogonalize(masks_path, output_path, epsilon=2)

Visualize results¶

In [ ]:

geoai.view_vector_interactive(output_path, tiles=test_raster_url)

geoai.view_vector_interactive(output_path, tiles=test_raster_url)

In [ ]:

geoai.create_split_map(
    left_layer=output_path,
    right_layer=test_raster_url,
    left_args={"style": {"color": "red", "fillOpacity": 0.2}},
    basemap=test_raster_url,
)

geoai.create_split_map( left_layer=output_path, right_layer=test_raster_url, left_args={"style": {"color": "red", "fillOpacity": 0.2}}, basemap=test_raster_url, )