tlseparation.scripts package

Submodules

tlseparation.scripts.automated_separation module

tlseparation.scripts.automated_separation.generic_tree(arr, knn_list=[40, 50, 80, 100, 120], voxel_size=0.05, retrace_steps=40)

Run an automated separation of a single tree point cloud.

Parameters:

arr : array

Three-dimensional point cloud of a single tree to perform the wood-leaf separation. This should be a n-dimensional array (m x n) containing a set of coordinates (n) over a set of points (m).

knn_lst: list

Set of knn values to use in the neighborhood search in classification steps. This variable will be directly used in a step containing the function reference_classification and its minimum and maximum values will be used in a different step with threshold_classification (both from classification.classify_wood). These values are directl dependent of point density and were defined based on a medium point density scenario (mean distance between points aroun 0.05m). Therefore, for higher density point clouds it’s recommended the use of larger knn values for optimal results.

verbose : bool

Option to set (or not) verbose output.

Returns:

wood_final : array

Wood point cloud.

leaf_final : array

Leaf point cloud.

tlseparation.scripts.automated_separation.large_tree_3(arr, class_file=[], knn_lst=[20, 40, 60, 80], gmm_nclasses=4, class_prob_threshold=0.95, cont_filt=True, cf_rad=None, verbose=False)

Run an automated separation of a single tree point cloud.

Parameters:

arr : array

Three-dimensional point cloud of a single tree to perform the wood-leaf separation. This should be a n-dimensional array (m x n) containing a set of coordinates (n) over a set of points (m).

class_file : str

Path to classes reference values file. This file will be loaded and its reference values are used to select wood and leaf classes.

knn_lst: list

Set of knn values to use in the neighborhood search in classification steps. This variable will be directly used in a step containing the function wlseparate_ref_voting and its minimum value will be used in another step containing wlseparate_abs (both from classification.wlseparate). These values are directly dependent of point density and were defined based on a medium point density scenario (mean distance between points aroun 0.05m). Therefore, for higher density point clouds it’s recommended the use of larger knn values for optimal results.

gmm_nclasses: int

Number of classes to use in Gaussian Mixture Classification. Default is 4.

cont_filt : boolean

Option to select if continuity_filter should be applied to wood and leaf point clouds. Default is True.

class_prob_threshold : float

Classification probability threshold to filter classes. This aims to avoid selecting points that are not confidently enough assigned to any given class. Default is 0.95.

cf_rad : float

Continuity filter search radius.

verbose : bool

Option to set (or not) verbose output.

Returns:

wood_final : array

Wood point cloud.

leaf_final : array

Leaf point cloud.

tlseparation.scripts.automated_separation.large_tree_4(arr, class_file=[], knn_lst=[20, 40, 60, 80], gmm_nclasses=4, class_prob_threshold=0.95, cont_filt=True, cf_rad=None, verbose=False)

Run an automated separation of a single tree point cloud.

Parameters:

arr : array

Three-dimensional point cloud of a single tree to perform the wood-leaf separation. This should be a n-dimensional array (m x n) containing a set of coordinates (n) over a set of points (m).

class_file : str

Path to classes reference values file. This file will be loaded and its reference values are used to select wood and leaf classes.

knn_lst: list

Set of knn values to use in the neighborhood search in classification steps. This variable will be directly used in a step containing the function wlseparate_ref_voting and its minimum value will be used in another step containing wlseparate_abs (both from classification.wlseparate). These values are directly dependent of point density and were defined based on a medium point density scenario (mean distance between points aroun 0.05m). Therefore, for higher density point clouds it’s recommended the use of larger knn values for optimal results.

gmm_nclasses: int

Number of classes to use in Gaussian Mixture Classification. Default is 4.

cont_filt : boolean

Option to select if continuity_filter should be applied to wood and leaf point clouds. Default is True.

class_prob_threshold : float

Classification probability threshold to filter classes. This aims to avoid selecting points that are not confidently enough assigned to any given class. Default is 0.95.

cf_rad : float

Continuity filter search radius.

verbose : bool

Option to set (or not) verbose output.

Returns:

wood_final : array

Wood point cloud.

leaf_final : array

Leaf point cloud.

tlseparation.scripts.automated_separation.nopath_generic_tree(arr, knn_list=[40, 50, 80, 100, 120])

Run an automated separation of a single tree point cloud.

Parameters:

arr : array

Three-dimensional point cloud of a single tree to perform the wood-leaf separation. This should be a n-dimensional array (m x n) containing a set of coordinates (n) over a set of points (m).

knn_lst: list

Set of knn values to use in the neighborhood search in classification steps. This variable will be directly used in a step containing the function reference_classification and its minimum and maximum values will be used in a different step with threshold_classification (both from classification.classify_wood). These values are directl dependent of point density and were defined based on a medium point density scenario (mean distance between points aroun 0.05m). Therefore, for higher density point clouds it’s recommended the use of larger knn values for optimal results.

Returns:

wood_final : array

Wood point cloud.

leaf_final : array

Leaf point cloud.

tlseparation.scripts.post_processing module

tlseparation.scripts.post_processing.isolated_clusters(arr, voxel_size=0.05, size_threshold=0.3, feature_threshold=0.6, min_pts=10)

Performs a connected component analysis to cluster points from a point cloud and filters them these clusters based on size and shape (geometric feature).

Parameters:

arr : array

Three-dimensional (m x n) array of a point cloud, where the coordinates are represented in the columns (n) and the points are represented in the rows (m).

voxel_size: float

Distance used to generate voxels from point cloud in order to perform the connected component analysis in 3D space.

size_threshold : int/float

Minimum size, on any dimension, for a cluster to be set as valid (True)

feature_threshold : float

Minimum feature value for the cluster to be set as elongated (True).

min_pts : int

Minimum number of points for the cluster to be set as valid (True).

Returns:

filter_mask : array

1D mask array setting True for valid poins in ‘arr’ and False otherwise.

Module contents