KJNodes/imagegenerated

Image Noise Augmentation(ImageNoiseAugmentation)

This node technically uses PyTorch to generate random noise that conforms to a normal distribution, and uniformly scales the noise intensity through the `noise_aug_strength` parameter.

Image Noise Augmentation

image
IMAGE
noise_aug_strength
FLOAT
seed
123
KJNodes

The Image Noise Augmentation node technically uses PyTorch to generate random noise conforming to a normal distribution and uniformly scales the noise intensity via the noise_aug_strength parameter.

The Image Noise Augmentation node is an image processing tool whose core function is to add controllable random noise to input images. By introducing noise, it alters the texture and details of an image, commonly used for data augmentation or to create specific input conditions for subsequent processing steps.

Node Function

This node technically uses PyTorch to generate random noise conforming to a normal distribution and uniformly scales the noise intensity via the noise_aug_strength parameter. It supports the standard ComfyUI IMAGE tensor format and utilizes the seed parameter to ensure the reproducibility of the generated noise pattern each time, which is crucial for workflows requiring stable output.

Node Parameter Description - Image Noise Augmentation

Connection Inputs (Inputs)

Parameter NameData TypeRequiredDefault ValueRange/OptionsDescription
imageIMAGEYes--image (Input Parameter): The source image to which noise will be added. Type is IMAGE.

Control Parameters (Parameters)

Parameter NameData TypeRequiredDefault ValueRange/OptionsDescription
noise_aug_strengthFLOATYesNone0.0 - 100.0 (Step: 0.001)noise_aug_strength (Input Parameter): Controls the intensity of the noise added to the image. The value range is from 0.0 to 100.0, with an adjustment step of 0.001.
seedINTYes1230 - 0xffffffffffffffff (Step: 1)seed (Input Parameter): The seed value used to control the random noise generation pattern, ensuring reproducible results. The value range is from 0 to 0xffffffffffffffff, with a step of 1.

Output

Parameter NameData TypeDescription
IMAGEIMAGEIMAGE (Output Parameter): The output image with added noise.

Usage Scenarios

In practical applications, you can connect this node after an image generation node to add some graininess to images that are too "clean" or smooth, simulating film texture or adding detail. It can also serve as part of data preprocessing to augment training datasets and improve the robustness of machine learning models.

The Image Noise Augmentation node is from the ComfyUI-KJNodes node package.

Comments

Sign in with GitHub to join the discussion.

Loading comments…