سال انتشار: ۱۳۹۰

محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر

تعداد صفحات: ۵

نویسنده(ها):

Seyyed Mohammad Sadegh Moosavi – n University of Science & Technology,
Ahmad Ayatollahi – Iran University of Science & Technology

چکیده:

A biometric system provides automatic identificationof an individual based on unique features or characteristicspossessed by the individual. Iris recognition is regarded as themost reliable and accurate biometric identification systemavailable. The pattern of the human iris differs from person toperson. The most important step in the iris recognition system isthe segmentation process. Different methods of irissegmentation have been presented so far. In this paper a noveltype of active contour has been developed which can beemployed in iris segmentation. In fact, the proposed method is acomposition of the greedy snake algorithm and the balloonactive contour. The novel active contour has been named greedy- balloon. Both of the greedy and balloon active contours havemore advantages than traditional active contour. That is whywe used this active contour for our research. The proposedalgorithm was tested on the CASIA Iris Image Database 1.0 andthe obtained results showed that the proposed method has anacceptable accuracy.