LoadImage

class deeptrack.features.LoadImage(path: str | List[str] | Callable[[...], str | List[str]], load_options: dict | Callable[[...], dict] | None = None, as_list: bool | Callable[[...], bool] = False, ndim: int | Callable[[...], int] = 3, to_grayscale: bool | Callable[[...], bool] = False, get_one_random: bool | Callable[[...], bool] = False, **kwargs: Dict[str, Any])

Bases: Feature

Load an image from disk.

This feature attempts to load an image file using a series of file readers (imageio, numpy, Pillow, and OpenCV) until a suitable reader is found. Additional options allow for converting the image to grayscale, reshaping it to a specified number of dimensions, or treating the first dimension as a list of images.

Parameters

pathPropertyLike[Union[str, List[str]]]

The path(s) to the image(s) to load. Can be a single string or a list of strings.

load_optionsPropertyLike[Dict[str, Any]], optional

Options passed to the file reader. Defaults to None.

as_listPropertyLike[bool], optional

If True, the first dimension of the image will be treated as a list. Defaults to False.

ndimPropertyLike[int], optional

Ensures the image has at least this many dimensions. Defaults to 3.

to_grayscalePropertyLike[bool], optional

If True, converts the image to grayscale. Defaults to False.

get_one_randomPropertyLike[bool], optional

If True, extracts a single random image from a stack of images. Only used when as_list is True. Defaults to False.

Raises

IOError

If no file reader could parse the file or the file does not exist.

Methods Summary

get(*ign, path, load_options, ndim, ...)

Load the image(s) from disk and process them.

Methods Documentation

get(*ign: Any, path: str | List[str], load_options: Dict[str, Any] | None, ndim: int, to_grayscale: bool, as_list: bool, get_one_random: bool, **kwargs: Dict[str, Any]) → ndarray

Load the image(s) from disk and process them.

Parameters

pathUnion[str, List[str]]

The path(s) to the image(s) to load.

load_optionsOptional[Dict[str, Any]]

Options passed to the file reader.

ndimint

Ensures the image has at least this many dimensions.

to_grayscalebool

Whether to convert the image to grayscale.

as_listbool

Whether to treat the first dimension as a list.

get_one_randombool

Whether to extract a single random image from a stack.

**kwargsDict[str, Any]

Additional keyword arguments.

Returns

np.ndarray

The loaded and processed image(s).

Raises

IOError

If no file reader could parse the file or the file does not exist.