convert_to_images#

class ImageType(value)[source]#

Image conversion types.

FULL = (False, False, False, False)#

Image with TCZYX dimensions.

FULL_BINARY = (True, False, False, False)#

Binary image with TCZYX dimensions.

FULL_BY_FRAME = (False, True, False, False)#

Image with TCZYX dimensions separated by frame.

FULL_BINARY_BY_FRAME = (True, True, False, False)#

Binary image with TCZYX dimensions separated by frame.

FLAT_BY_FRAME = (False, True, True, False)#

Image array flattened to YX dimensions separated by frame.

FLAT_BINARY_BY_FRAME = (True, True, True, False)#

Binary array flattened to YX dimensions separated by frame.

FLAT_RGBA_BY_FRAME = (False, True, True, True)#

RGBA array flattened to YX dimensions separated by frame .

convert_to_images(series_key: str, locations_tar: tarfile.TarFile, frame_spec: tuple[int, int, int], regions: list[str], box: tuple[int, int, int], chunk_size: int, image_type: ImageType) → list[tuple[int, int, np.ndarray, int | None]][source]#

Convert data to image arrays.

Images are extracted from lists of voxels. The initial converted image has dimensions in TCZYX order, such that T encodes the specified frames and C encodes the regions. The initial converted image is then further processed based on selected image type.

Parameters:

  • series_key – Simulation series key.

  • locations_tar – Archive of location data.

  • frame_spec – Specification for image frames.

  • regions – List of region channels.

  • box – Size of bounding box.

  • chunk_size – Size of each image chunk.

  • image_type – Image conversion type.

Returns:

List of image chunks, chunk indices, and frames.

split_array_chunks(array: ndarray, chunk_size: int) → list[tuple[int, int, numpy.ndarray]][source]#

Split arrays into smaller chunks.

Parameters:

  • array – Image array (dimensions in TCZYX order).

  • chunk_size – Size of each image chunk.

Returns:

List of array chunks and their relative indices.

flatten_array_chunk(array: ndarray, image_type: ImageType) → ndarray[source]#

Flatten array chunk along z axis.

When flattening to an RGBA array, each object is encoded as a unique color such that the object ID = R + G*256 + B*256*256 - 1 and background pixels are black (R = 0, G = 0, B = 0).

Parameters:

  • array – Image array (dimensions in TCZYX order).

  • image_type – Image conversion type.

Returns:

Flattened image array.