Description:
Currently, the most advanced spatial filtering logic relies on static anchor points (LineString) manually defined for the 'Northern France' region. While this method successfully reduced API costs by 90% replacing a standard rectangular grid (was particularly effective given the diagonal orientation of the Northern French coast, where a traditional bounding box would over-scrape inland areas), it can clearly be improved to rely less on human input.
Objective:
Develop a module capable of taking a geographical entity as input (e.g., 'Brittany' or a pair of coordinates) and automatically extracting the simplified coastline via Shapefiles or the OpenStreetMap API. The optimal efficiency/complexity ratio must be figured out as well.
Technical Challenges:
- Geometry Simplification: Found out the Douglas-Peucker algorithm would reduce the complexity of the stored string, I need to understand the way it works
- Complex Topology: Handling islands, peninsulas, and estuaries to avoid scraping 'empty' sea or deep inland areas.
- Automation: Auto-generating the 10km buffer zone immediately after extraction to streamline the scraping pipeline.
Description:
Currently, the most advanced spatial filtering logic relies on static anchor points (LineString) manually defined for the 'Northern France' region. While this method successfully reduced API costs by 90% replacing a standard rectangular grid (was particularly effective given the diagonal orientation of the Northern French coast, where a traditional bounding box would over-scrape inland areas), it can clearly be improved to rely less on human input.
Objective:
Develop a module capable of taking a geographical entity as input (e.g., 'Brittany' or a pair of coordinates) and automatically extracting the simplified coastline via Shapefiles or the OpenStreetMap API. The optimal efficiency/complexity ratio must be figured out as well.
Technical Challenges: