I am using OSMnx to get clean intersections from OpenStreetMaps road network. The intersection nodes are currently in (x,y) coordinates, but I want to plot them using lat lon coordinates.
From the example Jupiter notebook, OSMnx Example #14 Clean Intersection Cluster Nodes, I am able to get the street network and call ox.clean_intersections to produce the clean intersections.
import osmnx as ox, matplotlib.pyplot as plt, numpy as np
ox.config(use_cache=True, log_console=True)
%matplotlib inline
# get a street network and plot it with all edge intersections
address = '2700 Shattuck Ave, Berkeley, CA'
G = ox.graph_from_address(address, network_type='drive', distance=750)
G_proj = ox.project_graph(G)
# clean up the intersections and extract their xy coords
intersections = ox.clean_intersections(G_proj, tolerance=15, dead_ends=False)
points = np.array([point.xy for point in intersections])
I get a Panda Geoseries of the intersections, which looks like this:
0 POINT (564152.437121744 4189596.945341664)
1 POINT (564846.6779513165 4189615.534235776)
2 POINT (564571.2116373706 4189601.780093061)
Since the clean intersections are comprised of the centroids of clusters of merged nodes, they don't correspond to any particular node with an osm_id (which have lat lon coordinates).
How can I convert these (x,y) points to lat lon coordinates?
You projected the graph to meters to clean the intersections with a sensible tolerance parameter. Now you just need to project the cleaned intersection centroids back to lat-long:
import geopandas as gpd
gdf = gpd.GeoDataFrame(geometry=intersections)
gdf.crs = G_proj.graph['crs']
ox.project_gdf(gdf, to_latlong=True)