ngageoint/hootenanny

What it solves

Hootenanny is designed to solve the problem of map data conflation—merging multiple geospatial datasets into a single, seamless map. It automates the process of combining features from different sources while maintaining the provenance of the original data and ensuring the resulting map is up-to-date and routable.

How it works

The tool uses machine learning techniques and tailored conflation algorithms to match and merge geospatial features. It supports a wide range of feature types, including buildings, roads, railways, and rivers. For features that don't fit a specific category, it uses a generic geometry conflation approach. Users can define custom conflation algorithms via JavaScript or C++ and custom schema translations via JavaScript or Python to handle different data formats (such as OSM, Geonames, and TDS).

Who it’s for

This tool is for geospatial analysts and developers who need to combine large-scale topographic map data from multiple sources into a single authoritative source.

Highlights

• Automated and Semi-Automated Conflation: Combines features automatically or allows for manual review and resolution of uncertain matches.
• Extensible Architecture: Supports custom algorithms and schema translations via JS, Python, and C++.
• Multi-Format Support: Leverages OSM key-value pair tags to translate between various data schemas.
• Comprehensive Tooling: Includes a web UI based on Mapbox iD Editor, a Web Services API, and a Command Line Interface (CLI).
• Data Cleaning: Integrated with JOSM validation and auto-fix capabilities to correct erroneous data.