cars.applications.point_cloud_outlier_removal.small_components

this module contains the small_components point removal application class.

Module Contents

Classes

SmallComponents

SmallComponents

Functions

small_component_removal_wrapper(cloud, ...[, ...])

Small components outlier removal

epipolar_small_component_removal_wrapper(cloud, ...[, ...])

Small component outlier removal in epipolar geometry
class cars.applications.point_cloud_outlier_removal.small_components.SmallComponents(conf=None)[source]

Bases: cars.applications.point_cloud_outlier_removal.pc_out_removal.PointCloudOutlierRemoval

SmallComponents

check_conf(conf)[source]

Check configuration

Parameters

conf (dict) – configuration to check

Returns

overloaded configuration

Return type

dict

get_optimal_tile_size(max_ram_per_worker, superposing_point_clouds=1, point_cloud_resolution=0.5)[source]

Get the optimal tile size to use, depending on memory available

Parameters

  • max_ram_per_worker (int) – maximum ram available

  • superposing_point_clouds (int) – number of point clouds superposing

  • point_cloud_resolution (float) – resolution of point cloud

Returns

optimal tile size in meter

Return type

float

get_method()[source]

Get margins to use during point clouds fusion

Returns

algorithm method

Return type

string

get_on_ground_margin(resolution=0.5)[source]

Get margins to use during point clouds fusion

Returns

margin

Return type

float

run(merged_point_cloud, orchestrator=None, depth_map_dir=None, point_cloud_dir=None, dump_dir=None, epsg=None)[source]

Run PointCloudOutlierRemoval application.

Creates a CarsDataset filled with new point cloud tiles.

Parameters

  • merged_point_cloud (CarsDataset filled with pandas.DataFrame) –

    merged point cloud. CarsDataset contains:

    • Z x W Delayed tiles. Each tile will be a future pandas DataFrame containing:

      • data : with keys : “x”, “y”, “z”, “corr_msk” optional: “clr”, “msk”, “data_valid”,”coord_epi_geom_i”, “coord_epi_geom_j”,”idx_im_epi” - attrs with keys: “epsg”

    • attributes containing “bounds”, “ysize”, “xsize”, “epsg”

  • orchestrator – orchestrator used

  • depth_map_dir (str) – output depth map directory. If None output will be written in dump_dir if intermediate data is requested

  • point_cloud_dir (str) – output depth map directory. If None output will be written in dump_dir if intermediate data is requested

  • dump_dir (str) – dump dir for output (except depth map) if intermediate data is requested

  • epsg (int) – cartographic reference for the point cloud (array input)

Returns

filtered merged point cloud. CarsDataset contains:

  • Z x W Delayed tiles. Each tile will be a future pandas DataFrame containing:

    • data : with keys : “x”, “y”, “z”, “corr_msk” optional: “clr”, “msk”, “data_valid”,”coord_epi_geom_i”, “coord_epi_geom_j”,”idx_im_epi”

    • attrs with keys: “epsg”

  • attributes containing “bounds”, “ysize”, “xsize”, “epsg”

Return type

CarsDataset filled with xr.Dataset

cars.applications.point_cloud_outlier_removal.small_components.small_component_removal_wrapper(cloud, connection_distance, nb_points_threshold, clusters_distance_threshold, save_by_pair: bool = False, point_cloud_csv_file_name=None, point_cloud_laz_file_name=None, saving_info=None)[source]

Small components outlier removal

Parameters

  • cloud (pandas DataFrame) – cloud to filter

  • connection_distance (float) – connection distance

  • nb_points_threshold (int) –

  • clusters_distance_threshold (float) –

  • save_by_pair (bool) – save point cloud as pair

  • point_cloud_csv_file_name (str) – write point cloud as CSV in filename (if None, the point cloud is not written as csv)

  • point_cloud_laz_file_name (str) – write point cloud as laz in filename (if None, the point cloud is not written as laz)

  • saving_info (dict) – saving infos

Returns

filtered cloud

Return type

pandas DataFrame

cars.applications.point_cloud_outlier_removal.small_components.epipolar_small_component_removal_wrapper(cloud, connection_distance, nb_points_threshold, clusters_distance_threshold, half_epipolar_size, window, overlap, epsg, point_cloud_csv_file_name=None, point_cloud_laz_file_name=None, saving_info_epipolar=None, saving_info_flatten=None)[source]

Small component outlier removal in epipolar geometry

Parameters

  • epipolar_ds (xr.Dataset) – epipolar dataset to filter

  • connection_distance (float) – minimum distance of two connected points

  • nb_points_threshold (int) – minimum valid cluster size

  • clusters_distance_threshold (float) – max distance between an outlier cluster and other points

  • half_epipolar_size (int) – half size of the window used to search neighbors

  • window (list) – window of base tile [row min, row max, col min col max]

  • overlap (list) – overlap [row min, row max, col min col max]

  • epsg (int) – epsg code of the CRS used to compute distances

Returns

filtered dataset

Return type

xr.Dataset