1
$\begingroup$

I am trying to isolate (or segment to be precise) the vein pattern from Infra red image of a finger (using MATLAB) like shown below: enter image description here

So my output will be a binary image like (not the exact dimension of image shown):

enter image description here

I tried histogram equalization to make the veins more visible, but it was not that successful. Any help to proceed further would be very helpful.

$\endgroup$
3
$\begingroup$

You might want to also try running a ridge filter on the dataset; something like the frangi filter might work.

A sample result is

enter image description here

NB the white border is due to the border in the image I pulled from this site

$\endgroup$
1
$\begingroup$

I tried using enhance contrast and histogram equalization on your image using FIJI and it seems to look pretty decent.enter image description here

You might still have some problems with segmenting due to the non-uniformity depending on the method you use. If so perhaps look at local/adaptive threshold or something that looks for edges/maxima such as watershed based approaches.

$\endgroup$
  • $\begingroup$ I got similar image results previously. But cannot isolate the vein pattern, not able to produce the binary image as I have shown. I have used Otsu thresholding. Let me see other options. $\endgroup$ – dexterdev Apr 29 '14 at 9:54

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.