Skip to content

PixelDropoutNoise & RandomPixelDropoutNoise Modules

Note

PixelDropoutNoise Module

The PixelDropoutNoise class applies pixel dropout noise to an image by randomly dropping out pixels based on a specified probability, simulating the removal of pixels.

Attributes

  • dropout_prob : float

    The probability of a pixel being dropped out. A higher value leads to more pixels being dropped.

  • pixel_dimensions : Tuple[float, float]

    The dimensions of the dropout pixels (width, height), controlling how many pixels are removed in blocks.

  • pixel_color : str

    The color of the dropped-out pixels, specified as a hex color code (e.g., #FFFFFF for white).

Methods

add_noise()

add_noise(self, image: Image) -> Image:
Public method that applies pixel dropout noise to the input image by dropping out pixels according to the specified probability, dimensions, and color.

  • Parameters:

    • image : Image

      The input image to which pixel dropout noise will be added.

  • Returns:

    • Image:

      The image with pixel dropout noise applied.

_pixeldropout_noise()

_pixeldropout_noise(self, image: Image) -> Image:
Internal method that creates a dropout mask, randomly selects pixels for removal, and applies the dropout to the image.

  • Parameters:

    • image : Image

      The image to which pixel dropout noise will be applied.

  • Returns:

    • Image:

      The noisy image with pixel dropout applied.

Usage Example

from PIL import Image
from iftg.noises import PixelDropoutNoise

# Open an image
image = Image.open('path/to/image.tif')

# Create a PixelDropoutNoise object with a dropout probability of 0.2 and pixel dimensions of (5, 5)
pixel_dropout_noise = PixelDropoutNoise(dropout_prob=0.2, pixel_dimensions=(5, 5), pixel_color='#FF0000')

# Apply pixel dropout noise to the image
noisy_image = pixel_dropout_noise.add_noise(image)

# Save the noisy image
noisy_image.save('path/to/noisy_image.tif')


RandomPixelDropoutNoise Module

The RandomPixelDropoutNoise class extends the functionality of PixelDropoutNoise by allowing random selection of the dropout probability and pixel dimensions from specified ranges.

Attributes

  • dropout_prob_range : Tuple[float, float]

    The range for random selection of the dropout probability. This controls the likelihood of pixels being dropped out.

  • pixel_dimensions_range : Tuple[float, float]

    The range for random selection of the pixel dimensions (width, height). Larger ranges will result in larger dropout blocks.

  • pixel_color : str

    The color of the dropped-out pixels, specified as a hex color code (e.g., #FFFFFF for white).

Methods

add_noise()

Note

Each call to add_noise() results in a different dropout pattern by selecting random values for the dropout probability and pixel dimensions, introducing variability into the noise application.

add_noise(self, image: Image) -> Image:
Public method that applies random pixel dropout noise to the input image by selecting random dropout probability and pixel dimensions within the specified ranges.

  • Parameters:

    • image : Image

      The input image to which random pixel dropout noise will be added.

  • Returns:

    • Image:

      The image with random pixel dropout noise applied.

Usage Example

from PIL import Image
from iftg.noises import RandomPixelDropoutNoise

# Open an image
image = Image.open('path/to/image.tif')

# Create a RandomPixelDropoutNoise object with dropout probability range (0.1, 0.3) and pixel dimensions range (3, 6)
random_pixel_dropout_noise = RandomPixelDropoutNoise(dropout_prob_range=(0.1, 0.3), pixel_dimensions_range=(3, 6), pixel_color='#00FF00')

# Apply random pixel dropout noise to the image
random_noisy_image = random_pixel_dropout_noise.add_noise(image)

# Save the randomly noisy image
random_noisy_image.save('path/to/random_noisy_image.tif')