GeoCoverage with Circles [private]

GeoCoverage with Circles

Summary

• The client has a script to collect data from Google Places within 50km radius circles, but circles of each country and geospatial data are not considered.
• I integrated a new functionality: I obtained all the polygons of the countries in the world. Given one of them, circles with a radius of 50km are generated to cover the entire area with minimal overlap.

Introduction

• Data Source: I obtained the polygon data using OpenStreetMap (OSM).
• Circles Generator: Given the polygon of a country, the script generates circles that perform SpatialJoin with the country’s surface.
• Minimal Overlap: Minimal overlap is achieved by offsetting the centroids by exactly half the distance between them.
• Circles Radius: The client requires circles with a radius of 50km, but the system works with any radius.
• Exported Data: The country’s polygon, centroids, and circles are exported in 3 different GeoJSON files.
• Plots: Visualizations shown in Figure 1 are added.

Project Goal

Find all the circles given any polygon, and for each of them, gather geographic data.

Images

Figure 1: Circles of 50km radius are shown for 4 arbitrary countries.

Technologies

• OpenStreetMap (OSM): Open-source community for geospatial data.
• GeoPandas: Used for processing polygons, circles, and centroids.
• Matplotlib: For data visualization.

Project Highlights

• The ability to generate all circles given a country was successfully delivered to the client.
• Minimal overlap was essential for the requirement, as Google Place charges for each API call.
• A simple mechanism for exporting and visualizing data is provided.