cars_mesh.core.texture

Texturing methods to project radiometric information over surfaces to provide a realistic rendering.

Module Contents

Functions

tile_norm(arr[, q_percent])	Normalize an image to a 8-bit image by recomputing the color dynamic
process_raster_by_tile(rio_infile, rio_outfile, ...[, ...])	Function that executes a user defined function using a tiling process.
preprocess_image_texture(img_path, bbox, output_dir[, ...])	Function that transforms a 16-bit tif image into an 8-bit jpg cropped to
generate_uvs(→ numpy.ndarray)	Function that computes the UVs coordinates
texturing(→ cars_mesh.tools.handlers.Mesh)	Function that creates the texture of a mesh.

cars_mesh.core.texture.tile_norm(arr: numpy.ndarray, q_percent: tuple | list | numpy.ndarray = None)

Normalize an image to a 8-bit image by recomputing the color dynamic

Parameters:

• arr (np.ndarray) – Image array

• q_percent ((2, ) list or tuple or np.ndarray) – Percentage of the respectively minimum and maximum values of the image from which to clip the dynamic

cars_mesh.core.texture.process_raster_by_tile(rio_infile, rio_outfile, fn_process_tile, col_off=0, row_off=0, tile_size=(1000, 1000), dst_bands=None, **kwargs)

Function that executes a user defined function using a tiling process.

Parameters:

• rio_infile (rasterio.io.DatasetReader) –

• rio_outfile (rasterio.io.DatasetWriter) –

• fn_process_tile (Function) –

• col_off (int) –

• row_off (int) –

• tile_size (tuple or list or np.ndarray) – (row, col)

• dst_bands (list or tuple or np.ndarray or None) – List of bands to handle in input and output file

• kwargs –

cars_mesh.core.texture.preprocess_image_texture(img_path: str, bbox: tuple | list | numpy.ndarray, output_dir: str, tile_size: tuple | list | numpy.ndarray = (1000, 1000))

Function that transforms a 16-bit tif image into an 8-bit jpg cropped to a bbox extent

Parameters:

• img_path (str) – Path to the TIF image to crop and normalize

• bbox (list or tuple or np.ndarray) – Bounding box extent for cropping: [top_left_col, top_left_row, bottom_right_col, bottom_right_row]

• output_dir (str) – Output directory path

• tile_size ((2, ) tuple or list or np.ndarray) – Tile size for the processing applied by tile for memory purpose (row, col)

cars_mesh.core.texture.generate_uvs(img_pts: tuple | list | numpy.ndarray, triangles: tuple | list | numpy.ndarray, bbox: tuple | list | numpy.ndarray, img_texture_size: tuple | list | numpy.ndarray) → numpy.ndarray

Function that computes the UVs coordinates

Parameters:

• img_pts ((N, 2) list or tuple or np.ndarray) – Equivalent image coordinates of ground point cloud data

• triangles ((M, 3) list or tuple or np.ndarray) – Mesh triangle indexes. The indexes must correspond to the ‘img_pts’ indexes.

• bbox ((4, ) list or tuple or np.ndarray) – List of coordinates for respectively the top left and bottom right corners in image frame and in the order (col, row)

• img_texture_size ((2, ) list or tuple or np.ndarray) – Size (col, row) of the image texture

Returns:

uvs – Normalized coordinates in the image texture of each triangle vertex. Normalization is done according to the image texture size and in [0, 1]

Return type:

np.ndarray

cars_mesh.core.texture.texturing(mesh: cars_mesh.tools.handlers.Mesh, cfg: dict) → cars_mesh.tools.handlers.Mesh

Function that creates the texture of a mesh.

Parameters:

• mesh (Mesh) – Mesh object

• cfg (dict:) –

  Configuration dictionary. The algorithm retrieves the following information:

  • output_dir: Str Output directory to write the new texture image.

  • rpc_path: Str Path to the xml rpc file.

  • tif_img_path: Str Path to the TIF image.

  • utm_code: int UTM code of the point cloud

Returns:

mesh – Mesh object with texture parameters

Return type:

Mesh