This demo is for research & educational use only — not a medical device. Read full disclaimer.

AI-Based Image Denoising

Upload an image for denoising
This research demonstration applies AI-based denoising to reduce common noise, such as Gaussian, Poisson, or mixed noise. Images with extreme blur or heavy artifacts may not be restored accurately. Images with any dimension larger than 1024 pixels will be resized while preserving the original aspect ratio. Please contact us for batch processing or support for higher-resolution images. Please upload only images that you are permitted to use.
Denoising strength
Adjust the denoising level from normal to more aggressive.
Level: 3
0.0s
Example Input
Noisy_image_sigma_50

Upload a noisy image.

  • Noise may be Gaussian, Poisson, speckle, or mixed
  • Noise can be removed from a wide variety of image types
  • Applicable to medical images such as OCT, OCTA, X-ray, and MRI

The current model is designed to remove noise while preserving essential image features. The example shown is an image from the CBSD68 dataset corrupted with Gaussian noise (σ = 50). Denoising strength can be selected from 1 to 5, corresponding to multiple denoising iterations.

Example Output
Example AIGS screening output

The output visualization provides:

  • A denoised output image processed at Denoising Strength Level 1.
  • Medical images or images with complex noise may require processing at Denoising Strength Levels greater than 1.

Results are provided for research use only when applied to medical images.