Image Denoising

The main challenge in digital image processing is to remove noise from the original image. We demonstrate two commonly used algorithms in image denoising:

• Non-local means algorithms
• Total variation minimization

For non-local means algorithm, we calculate the weights so as to understand how algorithm works. Additionally, we consider the case of variable illumination and show how Nl-means is susceptible to such cases. We also discuss one possible way to deal with images with variable illumination.

Future Work

• Show that using fourier filtering actually improves Nl-means
• Support for RGB images