We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
GPU Implementation of Sub-Iris Technique in Iris Recognition System.
- Authors
Shahrizan Jamaludin; Nasharuddin Zainal; W. Mimi Diyana W. Zaki
- Abstract
Iris recognition has become a widely popular biometric system. The stable textures and features of the human iris have made such biometric systems efficient and accurate for purposes of verification and identification. The term non-ideal iris refers to a situation in which the iris is occluded by noise, including reflections, eyelashes, eyelids and so on. Most current iris recognition algorithms assume that the iris is not occluded, which is less accurate. A method using only some parts of the iris may be suitable to deal with a non-ideal iris. The current application of iris recognition systems are plagued by weaknesses such as slow processing times, especially when dealing with many irises. In this study, a sub-iris recognition technique is proposed to deal with the non-ideal iris, while reducing execution time via an embedded system using a graphical processing unit (GPU). The experiment revealed that the proposed method was accurate and fast.
- Subjects
IRIS recognition; GRAPHICS processing units; BIOMETRIC eye scanning systems; EYELASHES; COMPUTER algorithms
- Publication
Pertanika Journal of Science & Technology, 2017, Vol 25, Issue S, p263
- ISSN
0128-7680
- Publication type
Article