cars.applications.dem_generation.rasterization

this module contains the dichotomic dem generation application class.

Module Contents

Classes

Rasterization

Rasterization

Functions

edit_transform(input_dem[, resolution, transform])

Change transform of an image

reverse_dem(input_dem)

Compute the opposite of a DEM :

downsample_dem(input_dem, scale)

Downsample median DEM with median resampling

launch_bulldozer(input_dem, output_dir, ...)

Launch bulldozer on a DEM to smooth it

compute_stats(diff)

Compute and display statistics of difference between two DEM :
class cars.applications.dem_generation.rasterization.Rasterization(conf=None)[source]

Bases: cars.applications.dem_generation.dem_generation.DemGeneration

Rasterization

check_conf(conf)[source]

Check configuration

Parameters

conf (dict) – configuration to check

Returns

overloaded configuration

Return type

dict

run(triangulated_matches_list, output_dir, geoid_path, dem_roi_to_use=None, initial_elevation=None, cars_orchestrator=None)

Run dichotomic dem generation using matches

Parameters

  • triangulated_matches_list (list(pandas.Dataframe)) – list of triangulated matches positions must be in a metric system

  • output_dir (str) – directory to save dem

  • geoid_path – geoid path

  • dem_roi_to_use – dem roi polygon to use as roi

Returns

dem data computed with mean, min and max. dem is also saved in disk, and paths are available in attributes. (DEM_MEDIAN_PATH, DEM_MIN_PATH, DEM_MAX_PATH)

Return type

CarsDataset

cars.applications.dem_generation.rasterization.edit_transform(input_dem, resolution=None, transform=None)[source]

Change transform of an image :param input_res: path of image :type input_dem: str :param resolution: image resolution for new transform to apply :type resolution: float :param transform: new transform to apply if resolution is not given :type transform: affine.Affine

cars.applications.dem_generation.rasterization.reverse_dem(input_dem)[source]

Compute the opposite of a DEM : Altitudes sign is changed

Parameters

input_dem (str) – path of DEM to reverse

cars.applications.dem_generation.rasterization.downsample_dem(input_dem, scale)[source]

Downsample median DEM with median resampling

Parameters

input_dem (str) – path of DEM to downsample (only one band)

cars.applications.dem_generation.rasterization.launch_bulldozer(input_dem, output_dir, cars_orchestrator, max_object_size)[source]

Launch bulldozer on a DEM to smooth it

Parameters

  • input_dem (str) – path of DEM to reverse

  • output_dir (str) – directory where bulldozer output is dumped

  • cars_orchestrator (Orchestrator) – orchestrator of the whole pipeline (used to get number of workers)

  • max_object_size (int) – bulldozer parameter “max_object_size”

cars.applications.dem_generation.rasterization.compute_stats(diff)[source]

Compute and display statistics of difference between two DEM : Minimum, median, percentiles and maximum

Parameters

diff (numpy.array) – altimetric difference between two DEM