cars.bundleadjustment

cars-bundleadjustment

Functions

matches_concatenation(matches_list, pairing, nb_decimals)

Concatenate matches computed by pair: for the second and

estimate_intersection_residues_from_matches(geomodels, ...)

Compute intersections from multiple matches and estimate the

aggregate_matches_by_cell(matches_dataframe, step, ...)

Aggregate matches with a step computing from matches density

plane_regression(points, values)

Deduce a plane fitting points / values

create_deformation_grid(images, regular_matches, ...)

Compute the deformation grid with a defined step

refine_rpc(geomodels, old_coordinates, new_coordinates)

Compute new rpc matching old and new coordinates

write_rpcs_as_geom(refined_rpcs, out_dir)

Write RPCs as geomfiles

new_rpcs_from_matches(sensors, config_directory, ...)

Main function of cars-bundleadjustement for new RPCs estimation:

cars_bundle_adjustment(conf, no_run_sparse[, ...])

cars-bundleadjustement main:

cli()

Command Line Interface

Module Contents

cars.bundleadjustment.matches_concatenation(matches_list, pairing, nb_decimals)[source]

Concatenate matches computed by pair: for the second and subsequent pairs, the first image must be included in the list of previous images.

cars.bundleadjustment.estimate_intersection_residues_from_matches(geomodels, matches_dataframe, ignored=None)[source]

Compute intersections from multiple matches and estimate the residues as the difference between the previous sensor position and the inverse location of the multiple views intersection.

cars.bundleadjustment.aggregate_matches_by_cell(matches_dataframe, step, min_matches)[source]

Aggregate matches with a step computing from matches density matches density: footprint divided by number of matches to deduce a “resolution

cars.bundleadjustment.plane_regression(points, values)[source]

Deduce a plane fitting points / values

cars.bundleadjustment.create_deformation_grid(images, regular_matches, interp_mode, step, aggregate_step)[source]

Compute the deformation grid with a defined step if interp_mode is True, nan is replaced by the row mean else the deformation is a plane deformation

cars.bundleadjustment.refine_rpc(geomodels, old_coordinates, new_coordinates)[source]

Compute new rpc matching old and new coordinates

cars.bundleadjustment.write_rpcs_as_geom(refined_rpcs, out_dir)[source]

Write RPCs as geomfiles

cars.bundleadjustment.new_rpcs_from_matches(sensors, config_directory, sparse_matching_directory, pairing=None, nb_decimals=0, min_matches=50, step=5, aggregate_matches=True, interp_mode=False)[source]

Main function of cars-bundleadjustement for new RPCs estimation: - Retrieve matches from pairs and concatenate - Estimate residues by inverse location - Compute new RPCs

cars.bundleadjustment.cars_bundle_adjustment(conf, no_run_sparse, output_format='yaml')[source]

cars-bundleadjustement main: - Launch CARS to compute homologous points (run sparse matching) - Compute new RPCs

cars.bundleadjustment.cli()[source]

Command Line Interface