[FEAT] Add Support for Raster Data in UrbanMapper

[soniacq]

💡 Feature Request

Welcome to the UrbanMapper feature request hub! 🎉

Hey there, before you jump in, take a quick sec to check out the
important stuff below—it’ll help us keep the magic flowing smoothly.

Important

Before you proceed, please confirm the following:

• I’ve checked the API docs and confirmed this feature doesn’t already exist (no duplicates, please!).
• I’ve scoured the existing issues to make sure this hasn’t been requested before.
• I get that feature requests are reviewed periodically and might take a little time—patience is key!

---

Describe the Feature Request

Add support for raster datasets within UrbanMapper.

Currently, UrbanMapper supports geospatial vector formats—Point, LineString, and Polygon geometries—via GeoJSON and Shapefile ingestion. However, many urban analytics workflows rely on raster data such as satellite imagery, digital elevation models, and urban heatmaps, which are not currently supported.

Enabling native raster handling would expand UrbanMapper’s analytical capabilities, particularly for applications in remote sensing, terrain modeling, land cover classification, and environmental monitoring.

Describe Preferred Solution

To integrate raster support into UrbanMapper, the following components are proposed:

• New Loader Module:
  Introduce a RasterLoader under the loaders/ module system. This loader should:

  • Support formats such as GeoTIFF, JPEG2000, and PNG (with accompanying world files if needed)
  • Output standardized layer objects compatible with the existing map interface and session loading mechanism

• Layer Management Integration:
  Extend the existing Layer Panel to support raster overlays with:

  • Toggling and reordering alongside vector layers
  • Opacity control and styling options (e.g., grayscale, color ramps, band selection)
  • Integration with Leaflet-based renderers (e.g., georaster-layer-for-leaflet): https://geotiff.github.io/georaster-layer-for-leaflet-example/examples/elevation.html

Describe Alternatives

In current workflows, raster data must be converted into vector format (e.g., via contouring or classification).

Related Code

Additional Context

Raster support will unlock powerful new use cases:

• Urban heat island analysis using thermal imagery
• Land cover classification via satellite or drone imagery
• Visualization of slope, elevation, and flood risk models
• Integration of third-party imagery such as NYS orthoimagery or OpenTopography DEMs

This enhancement aligns well with UrbanMapper’s goal of facilitating exploratory urban analytics in a lightweight, flexible platform.

If the feature request is approved, would you be willing to submit a PR?

• Yes
• No

(No worries if you’re new to this—we’ll guide you every step of the way! 🤝)

[simonprovost]

Brills! Happy to see this coming. Maybe https://github.com/OSGeo/GDAL from https://github.com/sacridini/Awesome-Geospatial could be of interest.

However, I'm not sure I understood the Layer Management Integration. Do you mean the urban layer module or something else?

Next, as for Output standardized layer objects compatible with the existing map interface and session loading mechanism would that be possible to simply export a raster into a geodataframe? Or is that not possible? maybe: https://pypi.org/project/rastertodataframe/

Cheers

[soniacq]

@simonprovost Yes, it is possible to export a raster into a GeoDataFrame—tools like rastertodataframe can facilitate this conversion. This can be useful for integrating raster data with vector-based workflows, such as performing spatial joins, overlays, or aggregations using platforms like GeoPandas or PostGIS.

However, this approach comes with trade-offs. Raster data has a regular grid structure that encodes spatial continuity and resolution. When converted to vector format, this structure is often lost, which may lead to increased data volume and potential loss of precision or analytical capability, especially for continuous or high-resolution rasters.

Therefore, it may be important to retain the original raster structure as a layer object for certain use cases. Conversion to a GeoDataFrame can be used selectively when integration with vector-based components is required. That said, these are just my initial thoughts—the solution will likely evolve and improve as we work through the implementation.

[simonprovost]

I see I see! Lost of precision when converting to GeoDataframe. I am curious of such an implementation; let me know, guys. :)

[simonprovost]

Tip

Preliminary work can be retrieved through #70 by @JUDITH-sketch