Please use this identifier to cite or link to this item:
|Title:||Robust iris localisation in challenging scenarios|
Hélder Filipe Oliveira
|Abstract:||The use of images acquired in unconstrained scenarios is giving rise to new challenges in the field of iris recognition. Many works in literature reported excellent results in both iris segmentation and recognition but mostly with images acquired in controlled conditions. The intention to broaden the field of application of iris recognition, such as airport security or personal identification in mobile devices, is therefore hindered by the inherent unconstrained nature under which images are to be acquired. The proposed work focuses on mutual context information from iris centre and iris limbic and pupillary contours to perform robust and accurate iris segmentation in noisy images. The developed algorithm was tested on the MobBIO database with a promising 96% segmentation accuracy for the limbic contour.|
|Appears in Collections:||CTM - Articles in International Journals|
Files in This Item:
|2.47 MB||Adobe PDF||View/Open Request a copy|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.