— new Buildings model and implementation

Geoalert platform
Published in
3 min readFeb 26, 2021

Remember we released Mapflow — our new app to work with Geoalert Platform Hopefully you’ve had a chance to try it. If you haven’t, let me convince you to try some new cool features we’ve just implemented into Mapflow:

1. A new “high-density housing” AI-mapping model is available. It improves buildings automatic mapping in areas with terraced and other similar densely standing buildings, such as in the Middle East, in Africa etc.

A fragmet of densely built-up area in Morocco

While this new model is still in experimental beta, the credit cost for its use is very low. So you can test it. In case you do “urban mapping” and see this model applicable to your area of interest — we’ll be happy to learn from your experiments and to share the results of ours.

2. We’ve improved Mapflow visualisation capabilities through the implementation of which is open-sourced by Uber. It is a simple but powerful tool for data scientists to explore and analyse geospatial datasets.

Now you can view your processing results in

In particular, you can change visualisation parameters such as fill, opacity, etc. (the user guide for can be found here)

Moreover, you can visualise buildings in 3D if you have enabled the “building_height” option for your processing.

3D building visualisation using

Playing around with Kepler I tried to visualise an extract from “Urban Mapping” — our building footprints product with population counts. For example, you can use “Hexbin” or “Grid” map types to sum population in adjustable grid and visualise it — even in 3D.

Population data for Moscow city and Moscow region visualised in

However, you should be aware that all the data is rendered on the client’s side which can be a challenge for your PC performance.
There are also other ways to preview and visualise data in before you download it for importing into any specialised GIS software — you can head over to our embedded map viewer or to which is capable for small datasets convesrion to the other geospatial data types. But be aware that may hang your browser for good if you have more than a few tens of megabytes of data.

Stay tuned for more updates! Tell us what you think if you have any questions / suggestions regarding or other Geoalert products.

In Geoalert we love Maps and Data science. We apply Machine learning for automated detection and mapping over global Earth Observation data. You can find links and contacts below:

- Mapflow Platform
- Mapflow documentation
- Facebook page
- Linkedin page
- Telegram
- Github page

- Contacts



Geoalert platform

We apply Machine learning to automated analysis over Earth observation data