geopandas指向多边形

时间:2018-01-04 14:48:43

标签: python pandas geopandas point-in-polygon

我有一个GeoDataFrame的多边形(~30)和一个GeoDataFrame的点(~10k)

我希望在我的GeoDataFrame of Points中创建30个新列(具有适当的多边形名称),如果该点存在于多边形中,则使用简单的布尔值True / False。

例如,多边形的GeoDataFrame是这样的:

id  geometry
foo POLYGON ((-0.18353,51.51022, -0.18421,51.50767, -0.18253,51.50744, -0.1794,51.50914))
bar POLYGON ((-0.17003,51.50739, -0.16904,51.50604, -0.16488,51.50615, -0.1613,51.5091))

Points的GeoDataFrame是这样的:

counter     points
   1     ((-0.17987,51.50974))
   2     ((-0.16507,51.50925))

预期产出:

counter          points        foo    bar
   1    ((-0.17987,51.50974))  False  False
   1    ((-0.16507,51.50925))  False  False

我可以通过以下方式手动执行此操作:

foo = df_poly.loc[df_poly.id=='foo']
df_points['foo'] = df_points['points'].map(lambda x: True if foo.contains(x).any()==True else False

但鉴于我有30个多边形,我想知道是否有更好的方法。 感谢任何帮助!

1 个答案:

答案 0 :(得分:10)

不太清楚你实际拥有什么样的数据结构。此外,您的所有预期结果都是假的,因此很难检查。假设GeoSeries和GeoDataFrames,我会这样做:

from shapely.geometry import Point, Polygon
import geopandas

polys = geopandas.GeoSeries({
    'foo': Polygon([(5, 5), (5, 13), (13, 13), (13, 5)]),
    'bar': Polygon([(10, 10), (10, 15), (15, 15), (15, 10)]),
})

_pnts = [Point(3, 3), Point(8, 8), Point(11, 11)]
pnts = geopandas.GeoDataFrame(geometry=_pnts, index=['A', 'B', 'C'])
pnts = pnts.assign(**{key: pnts.within(geom) for key, geom in polys.items()})

print(pnts)

这让我:

        geometry    bar    foo
A    POINT (3 3)  False  False
B    POINT (8 8)  False   True
C  POINT (11 11)   True   True