Non-local means (NL) is an image de-noising process based on non-local averaging of all the pixels in an image. In particular, the amount of weighting for a pixel is based on the degree of similarity between a small patch centered around that pixel and the small patch centered around the pixel being de-noised.

If compared with other well-known denoising techniques, such as the Gaussian smoothing model, the anisotropic diffusion model, the total variation denoising, the neighborhood filters and an elegant variant, the Wiener local empirical filter, the translation invariant wavelet thresholding, the NL-means method noise looks more like white noise.^[1]

See also

• Total variation denoising
• Anisotropic diffusion
• Signal Processing
• Digital Image Processing
• Noise reduction

References

1. ↑ http://www.stanford.edu/~slansel/tutorial/Papers/Non-Local%20Means/On%20Image%20Denoising%20Methods.pdf

External links

• Recent trends in denoising tutorial
• Non-local image denoising, with code and online demonstration