30 Day Map Challenge 2024 - Day 7: Vintage

For day seven , the theme Vintage:

Map something modern in a vintage aesthetic. Create a map that captures the look and feel of historical cartography but focuses on a contemporary topic. Use muted colors, fonts, and classic elements.

Data

I’m not the most talented person when it comes to creating aesthetically pleasing graphics or designs. So today I’m going to focus on some technical details again and take the chance to learn something new. Specifically, I will take a look at how images can be georeferenced in order to place them correctly in a coordinate system.

To do this, I first downloaded a nearly 100 years old map of Dresden (1926) from the Deutsche Fotothek.

Since the dataset should reflect “something modern”, I chose to visualize locations of charging stations for electric cars. This dataset is obtained from OpenStreetMap via the overpass turbo webtool, using the following query:

node[amenity=charging_station](51.0, 13.6, 51.1, 13.83);
out geom;

I exported the result as a GeoJSON file, which I will import later using geopandas.

Implementation

First, we will of course need some imports. Today, there is nothing new to it:

import geopandas as gpd
import matplotlib.pyplot as plt
import rasterio
from rasterio.plot import show

Next, we need to georeference the map image. Opening the file in a somewhat advanced image editor (I use GIMP), allows to find the pixel coordinates of some very remarkable features, like crossings, buildings, or special features of the landscape. The same features have to be located using a mapping service, like Google Maps. This allows to find the real-world coordinates of those points and setup a list of so-called ground control points, form which a transformation can be calculated. The rasterio library already implements this functionality:

gcps = [
    rasterio.control.GroundControlPoint(col= 307, row= 332, y=51.08368419427998, x=13.686551101853567),
    rasterio.control.GroundControlPoint(col= 151, row=1202, y=51.01389086947117, x=13.665240136707837),
    rasterio.control.GroundControlPoint(col=1273, row=1231, y=51.01168185051654, x=13.808524372827599),
    rasterio.control.GroundControlPoint(col=1340, row= 263, y=51.08922315122832, x=13.818753999574017),
]
transformation = rasterio.transform.from_gcps(gcps)

with rasterio.open("data/map_dresden_1926.jpg", "r") as image_file:
    image_raw = image_file.read()

Furthermore, we need to load the locations of all charging stations exported from overpass turbo:

gdf_charging_stations = gpd.read_file("data/osm/charging_stations_dresden.geojson")

With the image loaded, the transformation set up, as well as the charging station data, we can finally create a nice visualization based on the old map of Dresden:

fig, ax = plt.subplots(1, 1, figsize=(15, 15))

show(
    image_raw,
    ax=ax,
    transform=transformation,
)

gdf_charging_stations["geometry"].plot(
    ax=ax,
    markersize=256,
    color="black",
    marker="$⚡$",
)

ax.set_axis_off()
plt.show()