MATLAB Answers

0

image deblurring or noise removal

Asked by huseyin kara on 26 Jun 2012
i have an image which created by camera. i almost correct this image color with the real image. for example; for a pixel's RGB values were changed [23 212 122] to [34 240 100]. but after this process (least square matrix method) my image has some noise around the edges of my image. how can i get rid of this noise?

  3 Comments

Ryan
on 26 Jun 2012
do you have an example image (showing the 'noise')? there are different types of noise and corresponding filters that remove them best. A good place to start is medfilt2(Image, [m n]) where [m n] is a window size. The smaller it is, the better it preserves image edges, but less it removes noise. Median filtering specializes in the removal of salt and pepper noise. It looks in the [m n] area around an individual pixel and replaces that pixel with the median value of all of those around it in its neighborhood.
You'd have to show the image, because I don't know why the noise only shows up around the edge of the image. Is that a darker or brighter color than you used in your training set?
http://b1206.hizliresim.com/y/v/8tp0y.png
this image is after some color correction processes. after i applied to it a couple filters median,... , the image was not seen well i mean it is not clear as before the filters. i am beginner about the image processing and can it be normal to get this result?

Sign in to comment.

1 Answer

Answer by Stephen
on 26 Jun 2012

try doing the same process to a matrix of ones and using ./ to get rid of the vignetting effect. Probably, whatever you're doing forces matlab to fill in with zeros at the edges.

  0 Comments

Sign in to comment.