Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have images where there is a lot of "black" background (few shades of black). (In many images at least half the pixels are background). I need to get interest points from the image, but because the background is not smooth (although black) , I get a lot of "false" key points in the background. What filter/s will fit best to keep the details in the image but will smooth the background as much as possible ? I tried median filter, but the size needed to reduce background noise ([5 5]) was to strong for the small details in the image I wanted to keep sharp, same for Gaussian with very small sigma.


share|improve this question
Can you add some images here? – Mohammad Feb 18 '13 at 5:58
up vote 0 down vote accepted

For small noises erosion followed by dilation can help as the erosion will clear the noise and dilation will bring back the deleted boundaries.Though i have still not implemented this Erosion+dilation option myself on grayscale but may be they can help

share|improve this answer

You can reduce image size or use scale-invariant interest point detector which can deal with blurry images.

Please note that using some elaborate denoising algorithm may change each image in a slightly different way, thus possibly distorting the positions of your interest points.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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