modules.glitches package¶

modules.glitches.abstract_glitch(img: numpy.ndarray, area: List[int] = None, translation_x: List[int] = [0, 0], n_slices: int = 20) → numpy.ndarray¶

Applies a glitch based on a pattern of repetition of rectangles in the specified area

of the given image.

Parameters

• img (ImageType) – A numpy array representing an image

• translation_x (List[int]) – [min, max] range for generating random glitch rectangles width

• area (List[int]) – img area (x, y, w, h) to apply the filter, default to None (all image)

• n_slices (int) – nº of glitch ‘slices’ that will appear in the provided area, defaults to 20

Returns

Resulting image

Return type

ImageType

modules.glitches.cycle_glitch(img: numpy.ndarray, area: List[int], translation_x: List[int], n_slices: int = 20) → numpy.ndarray¶

Draws kind of ‘artificial’ glitch effect on the specified area of the provided image, by drawing different distortioned ‘slices’. This is like the glitch() but making the slices follow a cyclic distortion.

Parameters

• img (ImageType) – A numpy array representing an image

• translation_x (List[int]) – [min, max] range for generating random glitch slices width

• area (List[int]) – img area (x, y, w, h) to apply the filter, default to None (all image)

• n_slices (int) – nº of glitch ‘slices’ that will appear in the provided area, defaults to 20

Returns

Resulting image

Return type

ImageType

modules.glitches.glitch(img: numpy.ndarray, area: List[int] = None, translation_x: List[int] = [0, 0], face: bool = False, n_slices: int = 20) → numpy.ndarray¶

Draws kind of ‘artificial’ glitch effect on the specified area of the provided image, by drawing different distortioned ‘slices’.

Parameters

• img (ImageType) – A numpy array representing an image

• translation_x (List[int]) – [min, max] range for generating random glitch slices width

• area (List[int]) – img area (x, y, w, h) to apply the filter, default to None (all image)

• face (bool) – If True filter is applied to img face area (if exists), defaults to False

• n_slices (int) – nº of glitch ‘slices’ that will appear in the provided area, defaults to 20

Returns

Resulting image

Return type

ImageType

modules.glitches.glitch_bytes(img: str, intensity: float = 0.1) → numpy.ndarray¶

This is the classic random glitch filter for an image. It depends on the provided image extension and returns a ‘random’ result of glitching & deforming the image. So my suggestion is to test it multiple times with the same image to see the different kinds of possible results. For now, it works with .jpg, .jpeg, .png, .bmp, .gif and .tiff. I recommend you to highly reduce the intensity for large images.

Parameters

• img (str) – The filepath of the image

• intensity (float) – Image deformation intensity, defaults to 0.1

Returns

Glitched image

Return type

ImageType

modules.glitches.glitch_plasma(img: numpy.ndarray, kernel: Tuple[int] = 3, 3) → numpy.ndarray¶

Applies a color-sections based glitch to the given image.

Parameters

• img (ImageType) – A numpy array representing an image

• kernel (Tuple[int]) – Kernel to apply to the gaussian blur filter

Returns

Resulting image

Return type

ImageType

modules.glitches.multiply(img: numpy.ndarray, factor: int) → numpy.ndarray¶

not documented yet

modules.glitches.offset_rect(img, start_x: int, start_y: int, chunk_length: int, side: str) → numpy.ndarray¶

Select a rectangle with the provided data whithin the image and translates it left or right as desired.

Parameters

• img (ImageType) – A numpy array representing an image

• start_x (int) – x start position for the rectangle

• start_y (int) – y start position for the rectangle

• chunk_length (int) – rectangle width

• side (int) – ‘left’ or ‘right’ for the rectangle’s position

Returns

Resulting image

Return type

ImageType

modules.glitches.offset_rect_colorized(img: numpy.ndarray, area: List[int], channel: int = 1, randomize: bool = False) → numpy.ndarray¶

Takes the desired area of the provided image and makes a ‘colorized translation’ of it based on the provided color channel of the (0,0, area_w,area_h) selection of the image.

Parameters

• img (ImageType) – A numpy array representing an image

• area (List[int]) – img area (x, y, w, h) to apply the filter, default to None (all image)

• channel (int) – Color channel to modify. You can pass a range as an slice, defaults to 1

• randomize (bool) – If True, randomized the area offset, defaults to False

Returns

Resulting image

Return type

ImageType

modules.glitches.pixelize_glitch(img: numpy.ndarray, area: List[int], n_slices: int, gtype: str = 'random', by_pixel: bool = True, channel: int = None, random_slice_width_range: List[int] = [90, 220], random_color_range: List[int] = [15, 260], skip_slices_range: List[int] = None) → numpy.ndarray¶

Draws a pattern of colorized pixel or slice over the area of the provided image. It can be with random colors or based on image palette.

Parameters

• img (ImageType) – A numpy array representing an image

• area (List[int]) – img area (x, y, w, h) to apply the filter, default to None (all image)

• n_slices (int) – nº of glitch ‘rows’ that will appear in the provided area

• gtype (str) – ‘image_based’, ‘image_based_inv’, ‘image_based_rand’ or ‘random’, defaults to ‘random’

• by_pixel (bool) – If true, the colorize process will be applied to each pixel, if False it will be applied to slices of a determined width, defaults to True

• channel (int) – Color channel to modify. You can pass a range as an slice, defaults to None

• random_slice_width_range (List[int]) – min/max of the range for randomly choose each slice width, defaults to [90, 220]

• random_color_range (List[int]) – min/max of the range in which random colors will be generated, defaults to [15, 260]

• skip_slices_range (List[int]) – min/max of a range for randomly exclude some rows of this effect, defaults to None

Returns

Resulting image

Return type

ImageType

modules.glitches.pixelize_glitch_vanish(img: numpy.ndarray, sampling_factor: int) → numpy.ndarray¶

This is a particular glitch that applies a kind of ‘vanish’ effect to image content based on the sampling factor provided. The result depends of the sampling factor which sould be based on the image size.

Parameters

• img (ImageType) – A numpy array representing an image

• sampling_factor (int) – Represents the intensity of the ‘vanish’ applied to the image. As greater, more distorsion more ‘pixelized’ the image will look like

Returns

Resulting image

Return type

ImageType

modules.glitches.spilled_glitch(img: numpy.ndarray, area: List[int], start_pos: int, vertical: bool = False) → numpy.ndarray¶

Spills down or right the area up/left header pixel colors creating a cascade of colorized lines.

Parameters

• img (ImageType) – A numpy array representing an image

• area (List[int]) – img area (x, y, w, h) to apply the filter, default to None (all image)

• start_pos (int) – y/x point where to start spilling from, depending on vertical parameter

• vertical (bool) – If True, spilled is applied based on an x start pos, else is based on the y start_pos, defaults to False

Returns

Resulting image

Return type

ImageType