pc: Point cloud data utilities

moraine.api.pc (source). Every name is also moraine.<name> (import moraine as mr).

pc_hix

pc_hix(gix, shape: tuple) -> ndarray

Compute the hillbert index for point cloud data.

Parameters:

Name	Type	Description	Default
gix		grid index	required
shape	tuple	(nlines, width)	required

pc_gix

pc_gix(hix, shape: tuple) -> ndarray

convert hillbert index to grid index.

Parameters:

Name	Type	Description	Default
hix		hillbert index	required
shape	tuple	(nlines, width)	required

pc_sort

pc_sort(idx: ndarray, shape: tuple = None) -> ndarray

Get the indices that sort the input.

Parameters:

Name	Type	Description	Default
idx	ndarray	unsorted gix (2D) or hix(1D)	required
shape	tuple	(nline, width), faster if provided for grid index input	None

Returns:

Type	Description
ndarray	indices that sort input

pc2ras

pc2ras(idx: ndarray, pc_data: ndarray, shape: tuple)

convert point cloud data to original raster, filled with nan

Parameters:

Name	Type	Description	Default
idx	ndarray	gix or hix array	required
pc_data	ndarray	data, 1D or more	required
shape	tuple	image shape	required

ras2pc

ras2pc(ras, gix)

pc_union

pc_union(idx1: ndarray, idx2: ndarray, shape: tuple = None) -> tuple

Get the union of two point cloud dataset. For points at their intersection, prefer idx1 rather than idx2

Parameters:

Name	Type	Description	Default
idx1	ndarray	int array, grid index or hillbert index of the first point cloud, sorted, without duplicates	required
idx2	ndarray	int array, grid index or hillbert index of the second point cloud, sorted, without duplicates	required
shape	tuple	image shape, faster if provided for grid index input	None

Returns:

Type	Description
tuple	the union index idx index of the point in output union index that originally in the first point cloud inv_iidx index of the point in output union index that only exist in the second point cloud inv_iidx2 index of the point in the second input index that are not in the first input point cloud

pc_intersect

pc_intersect(idx1: ndarray, idx2: ndarray, shape: tuple = None) -> tuple

Get the intersection of two point cloud dataset.

Parameters:

Name	Type	Description	Default
idx1	ndarray	int array, grid index or hillbert index of the first point cloud, sorted, without duplicates	required
idx2	ndarray	int array, grid index or hillbert index of the second point cloud, sorted, without duplicates	required
shape	tuple	image shape, faster if provided for grid index input	None

Returns:

Type	Description
tuple	the intersect index idx index of the point in first point cloud index that also exist in the second point cloud index of the point in second point cloud index that also exist in the first point cloud

pc_diff

pc_diff(idx1: ndarray, idx2: ndarray, shape: tuple = None) -> tuple

Get the point cloud in idx1 that are not in idx2.

Parameters:

Name	Type	Description	Default
idx1	ndarray	int array, grid index or hillbert index of the first point cloud, sorted, without duplicates	required
idx2	ndarray	int array, grid index or hillbert index of the second point cloud, sorted, without duplicates	required
shape	tuple	image shape, faster if provided for grid index input	None

Returns:

Type	Description
tuple	the diff index idx index of the point in first point cloud index that do not exist in the second point cloud

Examples

Run when this manual is built; the output follows each snippet.

import numpy as np
import moraine as mr

shape = (8, 8)                                                   # (nlines, width) of the raster
gix1 = np.array([[0, 1], [2, 3], [5, 5]], dtype=np.int32)         # (azimuth, range) of three points
hix1 = mr.pc_hix(gix1, shape)                                    # hilbert index of each point
print(hix1)
order = mr.pc_sort(hix1)                                         # point clouds are kept in hilbert order
hix1, gix1 = hix1[order], gix1[order]

gix2 = np.array([[2, 3], [7, 0]], dtype=np.int32)
hix2 = np.sort(mr.pc_hix(gix2, shape))
hix, in1, in2, new2 = mr.pc_union(hix1, hix2)                    # sorted inputs, sorted union
print(hix, mr.pc_gix(hix, shape).tolist())
print(in1, in2, new2)                                            # where the points of each input went

[ 1 11 34]
[ 1 11 34 63] [[0, 1], [2, 3], [5, 5], [7, 0]]
[0 1 2] [3] [1]

ras = mr.pc2ras(hix, np.arange(len(hix), dtype=np.float32), shape)   # point data on the raster, nan elsewhere
print(ras)
print(mr.ras2pc(ras, mr.pc_gix(hix, shape)))                          # and back

[[nan  0. nan nan nan nan nan nan]
 [nan nan nan nan nan nan nan nan]
 [nan nan nan  1. nan nan nan nan]
 [nan nan nan nan nan nan nan nan]
 [nan nan nan nan nan nan nan nan]
 [nan nan nan nan nan  2. nan nan]
 [nan nan nan nan nan nan nan nan]
 [ 3. nan nan nan nan nan nan nan]]
[0. 1. 2. 3.]