cars.applications.dense_matching.dense_matching_algo

This module is responsible for the dense matching algorithms: - thus it creates a disparity map from a pair of images

Classes

LinearInterpNearestExtrap

Linear interpolation and nearest neighbour extrapolation

Functions

compute_disparity_grid(disp_range_grid, ...)

Compute dense disparity grids min and max for pandora

compute_disparity(→ Dict[str, xarray.Dataset])

This function will compute disparity.

Module Contents

cars.applications.dense_matching.dense_matching_algo.compute_disparity_grid(disp_range_grid, left_image_object, right_image_object, used_band, threshold_disp_range_to_borders)[source]

Compute dense disparity grids min and max for pandora superposable to left image

Parameters:

  • disp_range_grid (CarsDataset) – disp range grid with min and max grids

  • left_image_object (xr.Dataset) – left image

:return disp min map, disp_max_map :rtype np.ndarray, np.ndarray

cars.applications.dense_matching.dense_matching_algo.compute_disparity(left_dataset, right_dataset, corr_cfg, used_band=None, disp_min_grid=None, disp_max_grid=None, compute_disparity_masks=False, cropped_range=None, margins_to_keep=0, classification_fusion_margin=-1, texture_bands=None, filter_incomplete_disparity_range=True, classif_bands_to_mask=None) → Dict[str, xarray.Dataset][source]

This function will compute disparity.

Parameters:

  • left_dataset (xarray.Dataset) – Dataset containing left image and mask

  • right_dataset (xarray.Dataset) – Dataset containing right image and mask

  • corr_cfg (dict) – Correlator configuration

  • used_band (str) – name of band used for correlation

  • disp_min_grid (np ndarray) – Minimum disparity grid (if None, value is taken from left dataset)

  • disp_max_grid (np ndarray) – Maximum disparity grid (if None, value is taken from left dataset)

  • compute_disparity_masks (Boolean) – Activation of compute_disparity_masks mode

  • cropped_range (numpy array) – true if disparity range was cropped

  • margins_to_keep (int) – margin to keep after dense matching

  • classification_fusion_margin (int) – the margin to add for the fusion

  • classif_bands_to_mask (list of str / int) – bands from classif to mask

Returns:

Disparity dataset

class cars.applications.dense_matching.dense_matching_algo.LinearInterpNearestExtrap(points, values)[source]

Linear interpolation and nearest neighbour extrapolation

interp
extrap
__call__(*args)[source]