Is your feature request related to a problem?

A popular way to train various kinds of generative neural networks is to add gaussian noise to an image and have them restore the original.

Desired solution

• Method add_gaussian_noise on Image
• Parameter standard_deviation to adjust the standard deviation of the noise
• Raise error if standard_deviation is below 0

Possible implementation:

• Create a new image with the same dimensions as the original containing only the noise:

from PIL import Image as PILImage

img_noise = PILImage.effect_noise((image.width, image.height), standard_deviation)

• Blend it with the original image using Image.blend

Possible alternatives (optional)

No response

Screenshots (optional)

No response

Additional Context (optional)

No response