`pandora2d.refinement.interpolation`

This module contains functions associated to the interpolation method used in the refinement step.

Module Contents

Interpolation class allows to perform the subpixel cost refinement step

class pandora2d.refinement.interpolation.Interpolation(cfg: Dict)[source]

Interpolation class allows to perform the subpixel cost refinement step

static check_conf(cfg: Dict) → Dict[source]

Check the refinement configuration

static wrapper_interp2d(params: numpy.ndarray, func: scipy.interpolate.interp2d) → numpy.ndarray[source]: Unpack tuple of arguments from minimize to fit in interp2d :param params: points coordinates :type params: np.array :param func: interp2d scipy function :type func: scipy.interpolate.interpolate.interp2d :return: minimum of interp2d functions at points :rtype: np.array

compute_cost_matrix(p_args) → Tuple[float, float][source]: Process the interpolation and minimize of a cost_matrix :param cost_volumes: Dataset with 4D datas :type cost_volumes: xr.Dataset :param coords_pix_row: array from disp_min_row to disp_max_row :type coords_pix_row: np.array :param coords_pix_col: array from disp_min_col to disp_max_col :type coords_pix_col: np.array :param args_matrix_cost: 2D matrix with cost for one pixel (dim: dispy, dispx) :type args_matrix_cost: np.array :return: res: min of args_matrix_cost in 2D :rtype: Tuple(float, float)

refinement_method(cost_volumes: xarray.Dataset, pixel_maps: xarray.Dataset) → Tuple[numpy.ndarray, numpy.ndarray][source]: Compute refine disparity maps :param cost_volumes: Cost_volumes has (row, col, disp_col, disp_row) dimensions :type cost_volumes: xr.Dataset :param pixel_maps: dataset of pixel disparity maps :type pixel_maps: xr.Dataset :return: delta_col, delta_row: subpixel disparity maps :rtype: Tuple[np.array, np.array]