Dissolving/merging overlapping polygons in shapefile but retaining a list of original attribute values in Python/GeoPandas

dissolvegeopandasoverlaypython

I am looking for an answer to this question, for overlapping polygons, but in Python and preferably with GeoPandas.

Background

I have files with up to 90'000 polygons, see Tab. 1 in the Minial Working Example (MWE) attached. These polygons sometimes overlap to a large extent (see Fig. 1 in MWE attached, e.g. the polygons "One", "Two", and "Three" all overlap with each other.)

Aim

I would like to to get Tab. 2 in the MWE attached, i.e. merging/dissolving the polygons within one GeoDataFrame and keeping all polygon IDs that were merged/dissolved to create the new shapes. Shapes that do not overlap with anything should be preserved (see "No Overlap" in Tab. 2 in the MWE attached).

What I have tried

'''geopandas.overlay(gdf1, gdf1, 'how'='union')''' -> does not seem to give back merged shapes, but rather intersections, whether I set the 'how' to 'union' or 'intersection' (see here for details)
Using this idea with Shapely. Seems tedious to me and I would prefer a Geopandas approach if possible. Plus I also did not find a way to do ID preservation
Using this idea I retrieve the right shapes, but loose as my Polygon IDs, respectively only keep the first one

I am sure there must be an easy pythonic way, but I think I lack the terminology to find the solution.

Best Answer

You can groupby and specify an aggregate function for each field you want to keep.

I have four fields,

LAN_KOD (my group by field)

KOMMUNKOD which I want as a list/comma separated string

SomeFloat which should be summed

geometry which should be dissolved:

 import geopandas as gpd

df = gpd.read_file(r'C:\GIS\data\testdata\ak_riks_2.shp')

aggfunctions = {'KOMMUNKOD':lambda x: ','.join(str(y) for y in list(x)), #With just list as function, it would not export as a shapefile. I needed to convert it to string
                'SomeFloat':'sum', 
                'geometry': lambda x: x.unary_union}

df2 = df.groupby('LAN_KOD').agg(aggfunctions)
df3 = gpd.GeoDataFrame(df2, geometry='geometry') #df2 became a pandas dataframe by the grouping
df3.to_file(r'C:\GIS\data\testdata\ak_riks_3.shp')

Related Solutions

[GIS] Dissolving shapefile but retaining a list of original attribute value in QGIS

It is now more than one year from your post... From QGIS 2.16, new Aggregates functions became available. One of the new expression is concatenate().

Your Sample

concatenate()

Open the attribute table and create a new text field (with enough string length), with expression:

# Object_ID  -> O_IDs 
concatenate(  "Object_ID" , group_by:= "Class", concatenator:= ', ' )

# Signal_Sp  -> Signals
concatenate( to_string("Signal_Sp"), group_by:= "Class", concatenator:= ', ' )

Please note concatenate() takes only texts (strings), so I had to apply to_string() to "Signal_Sp" which was integer type...

After concatenate() your attribute table will become like below:

Now we can dissolve the polygon keeping original attribute data.

I will need to delete old fields (Object_ID and Signal_Sp) afterwards.

[GIS] geopandas dissolve overlapping polygons

Using this example GeoSeries:

s = geopandas.GeoSeries([Polygon([(0, 0), (0, 2), (2, 2), (2, 0)]), Polygon([(0, 1), (0, 3), (2, 3), (2, 1)]),Polygon([(1, 0), (1, 2), (3, 2), (3, 0)]), Polygon([(4, 4), (4, 6), (6, 6), (6, 4)])])

s.plot(alpha=0.5, cmap='Set1')

We could create a matrix indicating which geometries are overlapping:

In [55]: overlap_matrix = s.apply(lambda x: s.overlaps(x)).values.astype(int)

In [56]: overlap_matrix
Out[56]: 
array([[0, 1, 1, 0],
       [1, 0, 1, 0],
       [1, 1, 0, 0],
       [0, 0, 0, 0]])

And from that get the groups using scipy's connected components:

In [57]: from scipy.sparse.csgraph import connected_components

In [58]: connected_components(overlap_matrix)
Out[58]: (2, array([0, 0, 0, 1], dtype=int32))

In [59]: n, ids = connected_components(overlap_matrix)

And use dissolve based on those groups:

In [60]: df = geopandas.GeoDataFrame({'geometry': s, 'group': ids})

In [61]: res = df.dissolve(by='group')

In [62]: res
Out[62]: 
                                                geometry
group                                                   
0      POLYGON ((0 0, 0 1, 0 2, 0 3, 2 3, 2 2, 3 2, 3...
1                    POLYGON ((4 4, 4 6, 6 6, 6 4, 4 4))

In [63]: res.plot(cmap='Set1')

But note: I am certainly not sure if creating such a matrix is the most efficient way.
Further note, this would actually be a nice enhancement to geopandas to enable this somehow (or at least to have a good solution in the examples)

Best Answer

Related Solutions

[GIS] Dissolving shapefile but retaining a list of original attribute value in QGIS

[GIS] geopandas dissolve overlapping polygons

Related Question