wed.utils.image_utils module

This submodule contains a set of functions for editing the following image:

_images/marquee.png
wed.utils.image_utils.arr2pil(frame: numpy.ndarray[Any, numpy.dtype[numpy.uint8]]) → PIL.Image.Image[source]

Convert from frame (BGR npt.NDArray) to image (RGB Image.Image)

Parameters

frame (npt.NDArray[np.uint8]) – A BGR npt.NDArray.

Returns

A RGB Image.Image

Return type

Image.Image

wed.utils.image_utils.pil2arr(image: PIL.Image.Image) → numpy.ndarray[Any, numpy.dtype[numpy.uint8]][source]

Convert from image (RGB Image.Image) to frame (BGR npt.NDArray).

Parameters

image (Image.Image) – A RGB Image.Image

Returns

A BGR npt.NDArray.

Return type

npt.NDArray[np.uint8]

wed.utils.image_utils.cv2plot(frame: numpy.ndarray[Any, numpy.dtype[numpy.uint8]], ax: Optional[matplotlib.axes._axes.Axes] = None, isBGR: bool = True) → matplotlib.axes._axes.Axes[source]

Plot a image.

Parameters

  • frame (npt.NDArray[np.uint8]) – Input image.

  • ax ([type], optional) – An Axes instance. Defaults to None.

  • isBGR (bool, optional) – Whether frame is BGR (OpenCV format) or not. Defaults to True.

Returns

Return type

Axes

wed.utils.image_utils.draw_text_in_pil(text: str, ttfontname: str = '/Users/iwasakishuto/Github/portfolio/Wed-Downtown-OP/data/fonts/LeagueGothic-1.601/static/TTF/LeagueGothic-Regular.ttf', img: Optional[PIL.Image.Image] = None, xy: Tuple = (0, 0), img_size: Tuple = (640, 360), img_mode: str = 'RGB', bgRGB: Union[str, Tuple] = 'white', textRGB: Union[str, Tuple] = 'black', fontsize: int = 16, fontwidth: Optional[int] = None, fontheight: Optional[int] = None, line_height: Optional[int] = None, anchor: Optional[str] = None, spacing: int = 4, align: str = 'left', direction: str = 'ltr', features: Optional[str] = None, language: Optional[str] = None, stroke_width: int = 0, stroke_fill: Optional[Union[str, Tuple]] = None, embedded_color: bool = False, text_width: Optional[int] = None, wrap_text: bool = False, drop_whitespace: bool = False, ret_position: str = 'word') → PIL.Image.Image[source]

Draw text in pillow image (Image.Image).

Parameters

  • text (str) – Text to be drawn to img.

  • ttfontname (str, optional) – A filename or file-like object containing a TrueType font. Defaults to FONT_LEAGUEGOTHIC_PATH.

  • img (Optional[Image.Image], optional) – The image to draw in. If this argment is None, img will be created using img_size and bgRGB arguments. Defaults to None.

  • xy (Tuple, optional) – Where to write the text. This value means the coordinates of (x, y). Defaults to (0, 0).

  • img_size (Tuple, optional) – The image size. Defaults to (640, 360).

  • img_mode (str, optional) – Optional mode to use for color values. Defaults to "RGB".

  • bgRGB (Union[str, Tuple], optional) – The color of background image. Defaults to "white".

  • textRGB (Union[str, Tuple], optional) – The color of text. Defaults to "black".

  • fontsize (int, optional) – The font size. Defaults to 16.

  • fontwidth (Optional[int], optional) – The font width. (If not given, automatically calculated.) Defaults to None.

  • fontheight (Optional[int], optional) – The font height. (If not given, automatically calculated.) Defaults to None.

  • line_height (Optional[int], optional) – The line height. (If not given, automatically calculated.) Defaults to None.

  • anchor (Optional[str], optional) – The text anchor alignment. Defaults to None.

  • spacing (int, optional) – The number of pixels between lines. Defaults to 4.

  • align (str, optional) – Determines the relative alignment of lines. Defaults to "left".

  • direction (str, optional) – Direction of the text. Defaults to "ltr".

  • features (Optional[str], optional) – A list of OpenType font features to be used during text layout. Defaults to None.

  • language (Optional[str], optional) – Language of the text. Defaults to None.

  • stroke_width (int, optional) – The width of the text stroke. Defaults to 0.

  • stroke_fill (Optional[Union[str, Tuple]], optional) – Color to use for the text stroke. If not given, will default to the textRGB parameter. Defaults to None.

  • embedded_color (bool, optional) – Whether to use font embedded color glyphs (COLR, CBDT, SBIX). Defaults to False.

  • text_width (Optional[int], optional) – The length of characters in one line. Defaults to None.

  • wrap_text (bool, optional) – Whether to wrap text for multilines or not. Defaults to False.

  • drop_whitespace (bool, optional) – If True, whitespace at the beginning and ending of every line. Defaults to False.

  • ret_position (str, optional) – Type of the position of next text to be returned. Please choose from ["line", "word"]. Defaults to "word".

Returns

img with text drawn.

Return type

Image.Image

wed.utils.image_utils.image_conversion(frame: numpy.ndarray[Any, numpy.dtype[numpy.uint8]], method: str) → numpy.ndarray[Any, numpy.dtype[numpy.uint8]][source]

Convert image by method method

Parameters

  • frame (npt.NDArray[np.uint8]) – Input image (BGR).

  • method (str) – How to convert an image.

Returns

Converted image.

Return type

npt.NDArray[np.uint8]

>>> import cv2
>>> import matplotlib.pyplot as plt
>>> from wed.utils import (
...     ROTATING_SQUARE_IMAGE_PATH,
...     SUPPORTED_CONVERSION_METHODS,
...     cv2plot,
...     image_conversion,
>>> )
>>> frame = cv2.imread(ROTATING_SQUARE_IMAGE_PATH)
>>> num_methods = len(SUPPORTED_CONVERSION_METHODS)
>>> fig, axes = plt.subplots(ncols=num_methods, nrows=1, figsize=(6 * num_methods, 4))
>>> for ax, method in zip(axes, SUPPORTED_CONVERSION_METHODS):
...     ax = cv2plot(image_conversion(frame, method=method), ax=ax)
...     ax.set_title(method)
>>> fig.show()
_images/wed-utils-image_utils-1.png
wed.utils.image_utils.nega_conversion(frame: numpy.ndarray[Any, numpy.dtype[numpy.uint8]]) → numpy.ndarray[Any, numpy.dtype[numpy.uint8]][source]

Convert images to negative and positive.

Parameters

frame (npt.NDArray[np.uint8]) – Input image (BGR).

Returns

[description]

Return type

npt.NDArray[np.uint8]

wed.utils.image_utils.alpha_composite(bg: PIL.Image.Image, paste: PIL.Image.Image, box: Tuple[int, int] = (0, 0)) → PIL.Image.Image[source]

Paste the paste image to bg with considering alpha channel.

Parameters

  • bg (Image.Image) – Background Image.

  • paste (Image.Image) – Image to paste.

  • box (Tuple[int,int], optional) – Where to paste the paste image on bg image. Defaults to (0,0).

Returns

The alpha-composited image.

Return type

Image.Image