Thanks Justin for sharing!

I've downloaded and extracted the India.geojsonl on a webserver.

ref about .geojsonl : https://www.interline.io/blog/geojsonl-extracts/
-> pretty useful! One can loop through a huge file without having to load
it all into RAM.

top lines look like:
{"type": "Feature", "properties": {},"geometry": {"type":
"Polygon","coordinates": [[[83.06380515611697, 25.34167404697847],
[83.06380909901775, 25.341635591519122], [83.06386494585949,
25.341640268589657], [83.06386100295869, 25.341678724047526],
[83.06380515611697, 25.34167404697847]]]}}
{"type": "Feature", "properties": {},"geometry": {"type":
"Polygon","coordinates": [[[87.87555977691633, 22.397660095199],
[87.8754256865811, 22.397658576750985], [87.87542690191553,
22.397566835386712], [87.87556099225075, 22.39756835383578],
[87.87555977691633, 22.397660095199]]]}}
...
So we have just basic polygons, one for each building, no properties or
categorization.

Here are some ideas on what to do with this:

1. Postgresql DB:
- Load all of these into a PostGreSQL DB
- Setup an api that will take a lat/lon and give all shapes within 1km
radius
- Next possible api: send a bounding polygon and get all buildings in that

2. Split up by district or lower level admin boundaries:
- Load the admin boundaries into a Postgresql DB
- Loop through each line (ie each building)
- Find out which place it's in by ST_Within query
- Dump it into a separate .geojsonl for that place
- We've now split it into multiple smaller files that are more usable, can
be loaded up by OSM mappers etc.

--
Cheers,
Nikhil VJ
https://nikhilvj.co.in


On Wed, May 18, 2022 at 3:51 AM Justin <[email protected]> wrote:

> [image: lg.png]
>
> https://github.com/microsoft/GlobalMLBuildingFootprints
>
>
> --
> Datameet is a community of Data Science enthusiasts in India. Know more
> about us by visiting http://datameet.org
> ---
> You received this message because you are subscribed to the Google Groups
> "datameet" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/datameet/68c0ecc4-5b07-4eb9-a69d-ae6e5ad6832bn%40googlegroups.com
> <https://groups.google.com/d/msgid/datameet/68c0ecc4-5b07-4eb9-a69d-ae6e5ad6832bn%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>

-- 
Datameet is a community of Data Science enthusiasts in India. Know more about 
us by visiting http://datameet.org
--- 
You received this message because you are subscribed to the Google Groups 
"datameet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/datameet/CAH7jeuM_yRYLDRir7vYjT7JVoP-J%2BkWAEkbp_U84QSb82AX3aA%40mail.gmail.com.

Reply via email to