Seyyed Mohammad Talebi - Iran University of Technology
Ahmad Ayatollahi - Iran University of Technology,
Seyyed Mohammad Sadegh Moosavi - Iran University of Technology

Image segmentation is a tool which is widely used in many applications related to image processing, and it comprises one of the most difficult steps of the process. This matter becomes more significant when we try to use the algorithms presented in complicated systems, such as the identity recognition systems. Different methods of image segmentation have been presented so far. In this article, a twostep algorithm has been proposed for iris segmentation. In the first step, a median filter has been applied to eliminate the noise, and in the next step, the modified active contour has been used for iris segmentation. The proposed algorithm was tested on the CASIA image data base and the obtained results showed that the proposed method has an acceptable accuracy.

Iris recognition, segmentation, activecontour, biometrics