OpenStreetMap (OSM) API and OSMnx in Python

Author: Dr. Vanessa Bastos | Dr. Gorden Jiang
School: School of Earth and Environment | Te Kura Aronukurangi
Course: Web GIS and Geoinformatics - GEOG 324 / Spatial Data Science - GISC412
Date: 8/9/24

Introduction to OSM and OSMnx
Main Functions and Methods
Examples of API Usage
Useful Resources

1. Introduction to OSM and OSMnx

OpenStreetMap (OSM) is a global, editable map database that allows users to contribute and access spatial data. OSM is often used in GIS applications, urban planning, and transportation analysis. OSMnx is a Python library that simplifies the interaction with OSM data and provides tools for retrieving, analyzing, and visualizing geospatial networks, particularly for street networks.

OSMnx can be used for:
- Downloading street networks and geographical data.
- Geocoding (converting addresses to coordinates).
- Performing graph/network analysis.
- Routing and calculating shortest paths.
- Visualizing geographic data, such as streets, buildings, and natural features.

2. Main Functions and Methods

This section summarizes the key OSM APIs and methods used in Python via the OSMnx library.

2.1 Extracting Road Networks

To retrieve a street network graph for a specific area, you can use ox.graph_from_place() or ox.graph_from_bbox().

import osmnx as ox
 
# Extract street network by place name
place = "University of Canterbury"
network = ox.graph_from_place(place)
 
# Extract street network using bounding box coordinates
north, south, east, west = -43.5199, -43.5282, 172.5883, 172.5729
network = ox.graph_from_bbox(north, south, east, west, network_type='all')
 
# Visualize the network
fig, ax = ox.plot_graph(network)

2.2 Geocoding with OSMnx

Geocoding converts an address to geographical coordinates. The ox.geocode() method is used to perform geocoding.

import geopandas as gpd
from shapely.geometry import Point
 
# Convert address to coordinates
address_list = ["11A Montana Avenue, Christchurch, New Zealand", ...]
geocode_result = ox.geocode(address_list[0])
 
# Store results as GeoDataFrame
gdf = gpd.GeoDataFrame({'Address': address_list}, geometry=geom, crs=4326)

2.3 Routing and Shortest Path

To compute the shortest path between two points on a network, you can use ox.distance.nearest_nodes() to find the nearest nodes and nx.shortest_path() to compute the route.

import networkx as nx
 
# Find nearest nodes to start and end points
start_point, end_point = (-43.5228, 172.5833), (-43.5215, 172.5798)
start_node = ox.distance.nearest_nodes(network, start_point[1], start_point[0])
end_node = ox.distance.nearest_nodes(network, end_point[1], end_point[0])
 
# Find the shortest path
route = nx.shortest_path(network, source=start_node, target=end_node, weight='length')
 
# Plot the route on the street network
ox.plot_graph_route(network, route)

2.4 Extracting Boundaries

To extract boundaries (such as the polygon of a city or neighborhood), ox.geocode_to_gdf() can be used.

# Extract a boundary polygon of a specified place
area = ox.geocode_to_gdf("University of Canterbury")
 
# Plot the boundary
area.plot()

2.5 Extracting Geographic Features by Tags

Tags in OSM describe the attributes of geographical features, such as roads, buildings, or natural features. You can use ox.features_from_place() to extract features by tags.

# Extract building features using the "building" tag
buildings = ox.features_from_place("University of Canterbury", tags={"building": True})
 
# Extract trees using the "natural" tag
trees = ox.features_from_place("University of Canterbury", tags={"natural": "tree"})
 
# Plot the extracted features
buildings.plot()
trees.plot()

2.6 Exporting Data from OSMnx

Once you have processed the data into a GeoDataFrame, you can export it to various formats such as Shapefiles or GeoPackages.

# Export data to a shapefile
buildings.to_file("Lab4_data/buildings.shp")
 
# Export data to a GeoPackage
buildings.to_file("Lab4_data/buildings.gpkg", driver="GPKG")

3. Examples of API Usage

Here are a few examples of using OSM APIs in a Pythonic way:

Extracting the road network for a specific area:
Retrieve the road network of a city or neighborhood using graph_from_place() and perform graph analysis or routing.
Geocoding multiple addresses:
Use ox.geocode() to geocode a list of addresses and store the results in a GeoDataFrame for further spatial analysis.
Visualizing shortest paths:
Combine NetworkX and OSMnx to compute and visualize the shortest path between two locations in a transportation network.
Exporting to standard GIS formats:
After data wrangling, export your data to formats like Shapefiles or GeoPackages for use in other GIS applications like QGIS or ArcGIS.

4. Useful Resources

OSMnx Documentation: OSMnx GitHub
OpenStreetMap Documentation: OSM Wiki
NetworkX Documentation: NetworkX Documentation
Christchurch City Council Spatial Open Data Portal: Christchurch Open Data

Hua Wang

Explorer

OSM