Georeferencing aligns images and scanned maps to real-world coordinates. Upload images without spatial information and use control points to position them accurately on the map. Once georeferenced, images overlay correctly with other spatial data.
Supported File Formats
Georeferencing works with image files: JPG, PNG, and PDF.
How Georeferencing Works
Georeferencing uses control points to align images to geographic coordinates. You place points on both the image and the map, and Atlas calculates the transformation needed to position the image correctly.
Control Points
Control points are matching locations visible on both your image and the reference map. Common control points include street intersections, building corners, landmarks, or other stable features that haven't changed over time.
Atlas requires at least 3 control points to georeference an image. More points improve accuracy, especially for distorted images like old paper maps. Spread points evenly across the image rather than clustering them in one area.
Georeference an Image
- Click Add data in the Layers panel
- Go to the Upload tab
- Upload your image file (JPG, PNG, or PDF)
- Click Georeference on the uploaded image layer
- Add control points by clicking matching locations on the image and map
- Add at least 3 points spread across the image
- Switch to Overlay mode to preview alignment
- Adjust control points if needed
- Click Confirm Georeferencing when satisfied
The image becomes a georeferenced layer aligned to real-world coordinates. It appears correctly positioned on the map and can be used with other spatial data.
Overlay Mode
Overlay mode shows your image positioned on the map based on control points. Use this mode to verify alignment before confirming georeferencing. Check that streets, buildings, and other features match between the image and the map.
If alignment looks incorrect, adjust control points by moving or deleting them and adding new ones. Switch back to georeferencing mode to make changes.
Best Practices
Use stable reference points that exist in both the image and the current map. Avoid temporary features, construction sites, or objects that may have moved.
Spread control points evenly across the image. Place points near corners and edges for better coverage, not just in the center.
Add more than 3 points for better accuracy. Use 6 or more control points for distorted images or when high precision is needed.
Choose clear, identifiable features as control points. Street intersections, building corners, and permanent landmarks work better than vague locations.
For distorted images like old paper maps, additional control points help correct warping and improve overall alignment.
Georeferenced Image Uses
Once georeferenced, images behave like standard GIS layers. Overlay them with current data to compare changes over time. Use them as basemaps for digitizing features like roads or boundaries. Adjust opacity and styling to blend with other layers. Export georeferenced images or share them with your team.
Troubleshooting
Poor alignment: Add more control points spread across the image. Verify that control points match the same locations on both the image and map.
Distorted results: Old maps often have paper distortion. Add 6 or more control points to correct warping, especially near edges and corners.
Can't find matching points: Use satellite imagery or different basemaps to find features visible in both the image and reference map.
Points won't place: Zoom in closer on both the image and map to place control points more precisely.
For more help with georeferencing issues, see the Troubleshooting Guide.