cars.devibrate

cars-devibrate: devibrate a high resolution DSM using a low resolution DSM

Attributes

package_path
GEOID_DEFAULT

Functions

acquisition_direction(sensor1, geomodel1, sensor2, ...)	Computes the mean acquisition of the input images pair
get_time_ground_direction(→ numpy.ndarray)	For a given image, compute the direction of increasing acquisition
project_coordinates_on_line(→ Union[float, numpy.ndarray])	Project coordinates (x, y) on a line starting from origin with a
lowres_initial_dem_splines_fit(...[, ext, order, ...])	This function takes 2 datasets containing DSM and models the
read_lowres_dsm(srtm_path, startx, starty, endx, endy)	Read an extract of the low resolution input DSM and return it as an Array
compute_splines(sensor1, geomodel1, sensor2, ...[, ...])	Compute a spline dict containing estimated splines, origin
cars_devibrate(used_conf, srtm_path, geoid_path[, ...])	Main fonction. Expects a dictionary from the CLI
cli()	Main cars-devibrate entrypoint (Command Line Interface)

Module Contents

cars.devibrate.package_path

cars.devibrate.GEOID_DEFAULT

cars.devibrate.acquisition_direction(sensor1, geomodel1, sensor2, geomodel2, geometry_plugin)

Computes the mean acquisition of the input images pair

Parameters:

• sensor1 – sensor image name of the first product

• geomodel1 – geomodel name of the first product

• sensor2 – sensor image name of the second product

• geomodel2 – geomodel name of the second product

Returns:

a tuple composed of : - the mean acquisition direction as a numpy array - the acquisition direction of the first product as a numpy array - the acquisition direction of the second product as a numpy array

cars.devibrate.get_time_ground_direction(sensor, geomodel, geometry_plugin, x_loc: float = None, y_loc: float = None, y_offset: float = None) → numpy.ndarray

For a given image, compute the direction of increasing acquisition time on ground. Done by two localizations at “y” and “y+y_offset” values.

Parameters:

• sensor – sensor image name

• geomodel – geomodel name

• x_loc – x location in image for estimation (default=center)

• y_loc – y location in image for estimation (default=1/4)

• y_offset – y location in image for estimation (default=1/2)

• dem – DEM for direct localisation function

• geoid – path to geoid file

Returns:

normalized direction vector as a numpy array

cars.devibrate.project_coordinates_on_line(x_coord: float | numpy.ndarray, y_coord: float | numpy.ndarray, origin: List[float] | numpy.ndarray, vec: List[float] | numpy.ndarray) → float | numpy.ndarray

Project coordinates (x, y) on a line starting from origin with a direction vector vec, and return the euclidean distances between projected points and origin.

Parameters:

• x_coord – scalar or vector of coordinates x

• y_coord – scalar or vector of coordinates y

• origin – coordinates of origin point for line

• vec – direction vector of line

Returns:

vector of distances of projected points to origin

cars.devibrate.lowres_initial_dem_splines_fit(lowres_dsm_from_matches: xarray.Dataset, lowres_initial_dem: xarray.Dataset, origin: numpy.ndarray, time_direction_vector: numpy.ndarray, ext: int = 3, order: int = 3, min_pts_per_time: int = 100, min_pts_along_time_direction: int = 100, butterworth_filter_order: int = 3, butterworth_critical_frequency: float = 0.05)

This function takes 2 datasets containing DSM and models the difference between the two as an UnivariateSpline along the direction given by origin and time_direction_vector. Internally, it looks for the highest smoothing factor that satisfies the rmse threshold.

Parameters:

• lowres_dsm_from_matches – Dataset containing the low resolution DSM obtained from matches, as returned by the rasterization.simple_rasterization_dataset function.

• lowres_initial_dem – Dataset containing the low resolution DEM, on the same grid as lowres_dsm_from_matches

• origin (list(float) or np.array(float) of size 2) – coordinates of origin point for line

• time_direction_vector (list(float) or np.array(float) of size 2) – direction vector of line

• ext – behavior outside of interpolation domain

• order – spline order

• min_pts_per_time – minimum number of points for each measurement

• min_pts_along_time_direction – minimum number of points for time direction

• butterworth_filter_order – Order of the filter. See scipy.signal.butter

• butterworth_critical_frequency – The filter critical frequency or frequencies. See scipy.signal.butter

cars.devibrate.read_lowres_dsm(srtm_path, startx, starty, endx, endy)

Read an extract of the low resolution input DSM and return it as an Array

cars.devibrate.compute_splines(sensor1, geomodel1, sensor2, geomodel2, matches, srtm_path, geoid_path, out_dir, min_pts_per_time: int = 100, min_pts_along_time_direction: int = 100, butterworth_filter_order: int = 3, butterworth_critical_frequency: float = 0.05)

Compute a spline dict containing estimated splines, origin and time_direction_vector

cars.devibrate.cars_devibrate(used_conf, srtm_path, geoid_path, min_pts_per_time: int = 100, min_pts_along_time_direction: int = 100, butterworth_filter_order: int = 3, butterworth_critical_frequency: float = 0.05)

Main fonction. Expects a dictionary from the CLI or directly the input parameters.

cars.devibrate.cli()

Main cars-devibrate entrypoint (Command Line Interface)