An approach towards iris localization for non cooperative images: A study
Priyanka Kaushal · 2010
Iris localization is the most important part of iris recognition which involves the detection of iris boundaries in an image. A very important need of this effective security system is to overcome the rigid constraints necessitated by the practical implementation of such a system. There are a few existing techniques for iris segmentation in which iris detection using Circular Hough Transform is the most reliable and popular and it has been implemented in this project. But there is a shortcoming in this technique. It does not perform well and does not gives high accuracy with images containing noise or occlusions caused by eyelids. Such kind of images constitute non cooperative data for iris recognition. To provide acceptable measures of accuracy, it is critical for an iris recognition system to overcome various noise effects introduced in images captured under different environment such as occlusions due to eyelids. This report discusses an approach towards less constraint iris recognition using occluded images. The Circular Hough Transform is implemented for few images and a novel approach towards iris localization and eyelids detection is studied..