imcircles to find iris region

Hi,

I was trying to segment iris region from the images in CASIA database( the images are noise free and of good quality). However hard i try, i couldn't detect the outer circle (boundary between iris and sclera). Though i was successful in getting pupil boundary. to get pupil boundary i used

     [centerspupil, radiipupil, metricpupil] = imfindcircles(A,[30 40], ...
    'ObjectPolarity','dark','Sensitivity',0.92,'EdgeThreshold',0.1);

Any suggestions??

[Elad]

can you upload an image?

[S]

I tried with images from CASIA database (which is publicly available), the inner circle is detected with a bit of tuning and there is no luck with the outer circle

Answer 1

[Image Analyst]

You need to play around with the values of the input parameters until you get something that works. Otherwise get the average radial profile going out from the pupil centroid and figure it out from that.

[S]

Thanks for your help,

but if we are trying to make a system which can practically segment iris region from any eye image (it may be from different database, images of different size) , this tuning of parameters won't be a good option (here, images from CASIA iris databse is of good quality and two prominent circles are there, still imfindcircles having a difficult time). Is there any other option which can find out the prominent circles present in an image?

and can you give me some more info about the radial profile and calculating outer circle from it?

[Image Analyst]

I suggest you first do some research on the papers where people have done it successfully. I'm sure they will have a much better algorithm than you or I could think up in a few minutes. Check out their papers here:

http://iris.usc.edu/Vision-Notes/bibliography/contentspeople.html#Face%20Recognition,%20Detection,%20Tracking,%20Gesture%20Recognition,%20Fingerprints,%20Biometrics

21.5.3 Biometrics, Iris Recognition Systems
    21.5.3.1 Iris Recognition Systems, Wavelet Based
    21.5.3.2 Iris Detection, Segmentation and Localization Systems
    21.5.3.3 Iris Recognition Systems, Systems, Evaluation, Comparison, Surveys