A Python tool for analyzing spatial patterns and proximity relationships of geographic entities using OpenStreetMap data. This project helps identify clusters, proximity networks, and spatial distributions of points of interest (POIs) in any given area.
- Entity Extraction: Retrieve geographic entities from OpenStreetMap based on custom tags
- Spatial Analysis: Analyze proximity relationships between entities using BallTree algorithm
- Visualization: Generate pie charts and proximity network visualizations
- Configurable Parameters: Customize search areas, entity types, walking distances, and more
- Brand Distribution: Count and analyze brand/entity distributions in a given area
- Retail Analysis: Analyze convenience store distributions and competitive proximity
- Urban Planning: Study spatial patterns of amenities, facilities, or services
- Market Research: Identify clusters and gaps in service coverage
- Accessibility Studies: Measure walking-distance relationships between locations
- Python 3.8 or higher
- pip package manager
- Clone the repository:
git clone https://github.com/MarlinZH/Geospatial_Patterns.git
cd Geospatial_Patterns- Install required dependencies:
pip install -r requirements.txtRun the example analysis:
python main.pyThis will analyze convenience stores in Shinjuku, Tokyo and generate:
- Entity count by brand
- Pie chart visualization
- Proximity network (stores within 2-minute walking distance)
from geospatial_analysis import GeospatialAnalyzer
# Initialize analyzer
analyzer = GeospatialAnalyzer(
area="Shinjuku, Tokyo",
tags={"shop": "convenience"},
walking_radius_meters=168 # ~2 minute walk
)
# Get entity counts
counts = analyzer.get_entity_counts()
# Get proximity relationships
proximity = analyzer.analyze_proximity()
# Generate visualizations
analyzer.visualize_distribution()# Analyze restaurants
analyzer = GeospatialAnalyzer(
area="Manhattan, New York",
tags={"amenity": "restaurant"}
)
# Analyze cafes
analyzer = GeospatialAnalyzer(
area="Paris, France",
tags={"amenity": "cafe"}
)
# Analyze pharmacies
analyzer = GeospatialAnalyzer(
area="London, UK",
tags={"amenity": "pharmacy"}
)# 5-minute walking radius (~420 meters at average walking speed)
analyzer = GeospatialAnalyzer(
area="Downtown Seattle",
tags={"shop": "supermarket"},
walking_radius_meters=420
)# Get entity list as DataFrame
entities_df = analyzer.get_entities_list()
entities_df.to_csv("entities.csv", index=False)
# Get proximity network
proximity_df = analyzer.analyze_proximity()
proximity_df.to_csv("proximity_network.csv", index=False)| Parameter | Type | Description | Default |
|---|---|---|---|
area |
str | Location name (searchable on OSM) | Required |
tags |
dict | OSM tag filters (e.g., {"shop": "convenience"}) |
Required |
walking_radius_meters |
int/float | Proximity radius in meters | 168 |
output_verbose |
bool | Print detailed outputs | False |
leaf_size |
int | BallTree leaf size for optimization | 15 |
Shops:
{"shop": "convenience"}- Convenience stores{"shop": "supermarket"}- Supermarkets{"shop": "bakery"}- Bakeries{"shop": "clothes"}- Clothing stores
Amenities:
{"amenity": "restaurant"}- Restaurants{"amenity": "cafe"}- Cafes{"amenity": "bank"}- Banks{"amenity": "pharmacy"}- Pharmacies{"amenity": "hospital"}- Hospitals
Transportation:
{"public_transport": "station"}- Transit stations{"amenity": "parking"}- Parking facilities
For more tags, visit the OSM Tag Wiki.
Brand
7-Eleven 45
FamilyMart 38
Lawson 32
NewDays 5
Name: count, dtype: int64
indices
0 [FamilyMart, Lawson]
1 [7-Eleven, NewDays, FamilyMart]
2 [7-Eleven]
...
Geospatial_Patterns/
├── main.py # Main script with example usage
├── geospatial_analysis.py # Core analysis class
├── example_configs.py # Pre-configured examples
├── requirements.txt # Python dependencies
├── README.md # This file
├── CONTRIBUTING.md # Contribution guidelines
├── LICENSE # MIT License
├── AI_ASSISTANT_PROMPT.md # AI assistant usage guide
├── .gitignore # Git ignore rules
└── cache/ # OSM data cache (gitignored)
- Geocoding: Uses OSMnx to convert location names to geographic boundaries
- Spatial Queries: Retrieves entities using OpenStreetMap Overpass API
- Proximity Search: Implements BallTree with Haversine metric for efficient nearest-neighbor search
- Coordinate System: Uses WGS84 (EPSG:4326) for geographic coordinates
The default walking radius of 168 meters represents approximately a 2-minute walk, assuming:
- Average walking speed: 1.4 m/s (5 km/h)
- Calculation: 1.4 m/s × 120 seconds = 168 meters
To customize: walking_radius_meters = speed_m_per_s × time_seconds
Need help with this project? Check out AI_ASSISTANT_PROMPT.md for ready-to-use prompts that give AI assistants the context they need to help you effectively.
Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.
See CONTRIBUTING.md for detailed guidelines.
- Fork the repository
- Create a feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
This project is open source and available under the MIT License.
- OSMnx - For OpenStreetMap data retrieval
- GeoPandas - For geospatial data manipulation
- Scikit-learn - For BallTree spatial indexing
- OpenStreetMap - For geographic data
Project Link: https://github.com/MarlinZH/Geospatial_Patterns
- Entity extraction relies on
brand:entag availability in OSM data - Large areas or dense urban regions may have slow query times
- Requires internet connection for OSM data retrieval
- Add interactive map visualizations with Folium
- Support for batch processing multiple areas
- Time-series analysis of entity changes
- Heat map generation for entity density
- CLI interface with argument parsing
- Web API for programmatic access
- Support for custom distance metrics
- Export to GeoJSON format
Note: This tool uses OpenStreetMap data. Please review the OSM Copyright and License for data usage terms.