I would like to know the basic differneces between Gaussian filter and Bilateral filter in digital image processing... I just know that both are used to smooth out an image but I would really appreciate if somebody let me know what is the purpose and scenario of each of these filters. Thanking in advance!


1 Answer 1


The Gaussian filter corresponds to isotropic diffusion (or the heat equation if you are familiar with it). Its application to an image will blur it uniformly, whatever the image content, edges, etc.

The bilateral filter is almost like a Gaussian filter, except that the Gaussian is modulated by a function that computes the similarity between the central pixel (where the filter is applied) and a pixel in its neighborhood (that is used in blurring).

  • If the two pixel values are very close, it multiplies the Gaussian coefficient by something close to 1, and hence it is equivalent to Gaussian filtering.
  • If the pixel values are very different , it will multiply the Gaussian coefficient by a number close to 0, thus turning off the Gaussian filtering for this pixel.

Intuitively, this behaviour yields the following result: Gaussian filtering in uniform areas of the image, no filtering across object borders. The bilateral filter will produce a more pleasant results, because it will avoid the introduction of blur between objects while still removing noise in uniform areas.

  • 1
    $\begingroup$ A bilateral filter is an "edge preserving filter" $\endgroup$
    – ijuneja
    Nov 9, 2018 at 19:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.