cartagen.smooth_gaussian

smooth_gaussian(geometry, sigma=30, sample=None, densify=True)

Smooth a line or a polygon and attenuate its inflexion points.

The gaussian smoothing has been studied by Babaud et al. [1] for image processing, and by Plazanet [2] for the generalisation of cartographic features.

Accept Multi geometries. If a polygon is provided, it also apply the smoothing to its holes using the same parameters.

Parameters:

• geometry (LineString, MultiLineString, Polygon, MultiPolygon, LinearRing) – The geometry to smooth. If a line is provided, the first and last vertexes are kept. If a polygon is provided, every vertex is smoothed.

• sigma (float, optional) – Gaussian filter strength. Default value to 30, which is a high value.

• sample (float, optional) – The length in meter between each nodes after resampling the geometry. If not provided, the sample is derived from the geometry and is the average distance between each consecutive vertex.

• densify (bool, optional) – Whether the resulting geometry should keep the new vertex density. Default to True.

Returns:

LineString, Polygon, MultiLineString, MultiPolygon

See also

smooth_platre

Smooth a line and preserve the integrity of sharp turns.

smooth_snake

Smooth a line or polygon using the snake method.

smooth_taubin

Smooth a line or polygon and prevent shrinkage.

smooth_topographic

Smooth a line or polygon and mimic hand-made cartographic generalization.

smooth_wma

Smooth a line or polygon using a low-pass filter.

References

[1]

Jean Babaud, Andrew P. Witkin, Michel Baudin, and Richard O. Duda. Uniqueness of the gaussian kernel for scale-space filtering. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8:26–33, 1986. URL: https://api.semanticscholar.org/CorpusID:18295906.

[2]

Corinne Plazanet. Enrichissement des bases de données géographiques : analyse de la géométrie des objets linéaires pour la généralisation cartographique (application aux routes). PhD thesis, École nationale des sciences géographiques, Marne-la-Vallée, France, 1996. Thèse de doctorat dirigée par Désarménien, Jacques Sciences de l’information géographique Marne-la-Vallée 1996. URL: http://www.theses.fr/1996MARN0001.

Examples

>>> line = LineString([(0, 0), (1, 1), (2, 0), (5, 3)])
>>> c4.smooth_gaussian(line, 1)
<LINESTRING (0 0, 1.666666666666667 0.6051115971014416, 3.333333333333334 1.6051115971014418, 5 3)>

>>> polygon = Polygon([(0, 0), (0, 1), (1, 1), (1, 0), (0, 0)])
>>> c4.smooth_gaussian(polygon, 1)
<POLYGON ((0.1168459780814714 0.3005282653219513, ... 0.1168459780814714 0.3005282653219513))>

(Source code)