Repository navigation
Enclosed tessellation post-processing #695
Description
Description / Summary
Add the enclosed tessellation post processing steps from : https://github.com/uscuni/urban_taxonomy/blob/84336d16afb968743e1ecf54553eefab7e8fa96e/src/core/generate_elements.py#L156
Value / benefit
This will fix the duplicated/split tessellation cells that sometimes result from enclosure deliniation
Implementation details
No response
Tasks to complete
No response
Activity
u3ks commentedon Jun 17, 2025
@martinfleis do you have any examples at hand ? I remember you used OSM data for Bristol ? Otherwise I gotta go through the cadastre and isolate some that way.
martinfleis commentedon Jun 17, 2025
At GISRUK, I used Bath - https://github.com/martinfleis/gisruk2025/blob/main/workshop.ipynb You may be able to get something using this code. I don't really have any specific example but my guess is that any decently large place will generate some.
u3ks commentedon Nov 2, 2025
I think this should be a separate function that returns the indices of:
I dont think it can be / should be a argument in momepy.enclosed_tessellation since then it may drop buildings silently. We can use the place from the morphotope/clustering tutorial as an example.
martinfleis commentedon Nov 2, 2025
I thought that the only issue we need to deal with are multipolygons of which one part lies under the building and the other is disjoint. Neither of your options sounds like that. Can you elaborate on issues you are trying to solve here?