View Metadata¶

Install Package

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

In [1]:

# %pip install geoai-py

# %pip install geoai-py

Import the package

In [2]:

import geoai

import geoai

Define URLs for sample datasets: a NAIP imagery raster file and a building footprints vector file

In [3]:

raster_url = (
    "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_train.tif"
)
vector_url = "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_train_buildings.geojson"

raster_url = ( "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_train.tif" ) vector_url = "https://huggingface.co/datasets/giswqs/geospatial/resolve/main/naip_train_buildings.geojson"

Download the raster file (NAIP imagery) and save it locally

In [4]:

raster_path = geoai.download_file(raster_url)

raster_path = geoai.download_file(raster_url)

File already exists: naip_train.tif

Download the vector file (building footprints) and save it locally

In [5]:

vector_path = geoai.download_file(vector_url)

vector_path = geoai.download_file(vector_url)

File already exists: naip_train_buildings.geojson

Display metadata about the raster file, including dimensions, resolution, projection, and bands

In [6]:

geoai.get_raster_info(raster_path)["band_stats"]

geoai.get_raster_info(raster_path)["band_stats"]

Out[6]:

[{'band': 1,
  'min': 12.0,
  'max': 251.0,
  'mean': 150.6730747259594,
  'std': 48.01908734374099},
 {'band': 2,
  'min': 49.0,
  'max': 251.0,
  'mean': 141.92468895229808,
  'std': 43.46595463573497},
 {'band': 3,
  'min': 53.0,
  'max': 251.0,
  'mean': 120.89909373405554,
  'std': 41.78086244480775},
 {'band': 4,
  'min': 22.0,
  'max': 251.0,
  'mean': 159.68995855062735,
  'std': 54.95588423977727}]

In [7]:

geoai.print_raster_info(raster_path, figsize=(18, 10))

geoai.print_raster_info(raster_path, figsize=(18, 10))

===== RASTER INFORMATION: naip_train.tif =====
Driver: GTiff
Dimensions: 2503 x 1126 pixels
Number of bands: 4
Data type: uint8
Coordinate Reference System: EPSG:26911
Georeferenced Bounds: BoundingBox(left=454780.8, bottom=5277567.0, right=456282.6, top=5278242.6)
Pixel Resolution: 0.5999999999999953, 0.5999999999996691
NoData Value: None

----- Band Statistics -----
Band 1:
  Min: 12.00
  Max: 251.00
  Mean: 150.67
  Std Dev: 48.02
Band 2:
  Min: 49.00
  Max: 251.00
  Mean: 141.92
  Std Dev: 43.47
Band 3:
  Min: 53.00
  Max: 251.00
  Mean: 120.90
  Std Dev: 41.78
Band 4:
  Min: 22.00
  Max: 251.00
  Mean: 159.69
  Std Dev: 54.96

Display metadata about the vector file, including geometry type, feature count, extent, and attributes

In [8]:

geoai.print_vector_info(vector_path, figsize=(18, 10))

geoai.print_vector_info(vector_path, figsize=(18, 10))

===== VECTOR INFORMATION: naip_train_buildings.geojson =====
Driver: GEOJSON
Feature count: 735
Geometry types: {'Polygon': 735}
Coordinate Reference System: EPSG:4326
Bounds: [-117.6017984, 47.65016239407519, -117.58246913360121, 47.655846]
Number of attributes: 3
Attribute names: id, height, class

----- Attribute Statistics -----
Attribute: id
  min: 1
  max: 735
  mean: 368.0000
  std: 212.3205
  null_count: 0
Attribute: height
  min: 0
  max: 0
  mean: 0.0000
  std: 0.0000
  null_count: 0

Analyze the "height" attribute of buildings to obtain statistical information

In [9]:

geoai.analyze_vector_attributes(vector_path, "height")

geoai.analyze_vector_attributes(vector_path, "height")

Out[9]:

{'attribute': 'height',
 'type': 'numeric',
 'count': np.int64(735),
 'null_count': np.int64(0),
 'min': np.int32(0),
 'max': np.int32(0),
 'mean': np.float64(0.0),
 'median': np.float64(0.0),
 'std': np.float64(0.0),
 'unique_values': 1}

Create a visualization of building footprints colored by their height values

In [10]:

geoai.visualize_vector_by_attribute(vector_path, "height")

geoai.visualize_vector_by_attribute(vector_path, "height")

Clip the raster file to a specified extent

In [11]:

clip_raster_path = "naip_clip.tif"
geoai.clip_raster_by_bbox(
    raster_path,
    clip_raster_path,
    bbox=(0, 0, 500, 500),
    bands=[1, 2, 3],
    bbox_type="pixel",
)

clip_raster_path = "naip_clip.tif" geoai.clip_raster_by_bbox( raster_path, clip_raster_path, bbox=(0, 0, 500, 500), bands=[1, 2, 3], bbox_type="pixel", )

Out[11]:

'naip_clip.tif'

In [12]:

geoai.view_image(clip_raster_path)

geoai.view_image(clip_raster_path)