الفهرس 
BiometricsOnly 14 pages are availabe for public view
 |  | from | 99 |  |  |
| | | from | 99 | | |
Abstract
Iris recognition systems have been widely deployed for authentication in many sensitive security areas for its accuracy and consistency. However, as the iris technology evolves, ways to attack it evolve too. Fake iris samples could be used to spoof the iris recognition system. As a result, Iris liveness detection methods have been developed. These methods read the users physiological signs of life to verify if the iris pattern acquired for identification is fake or real.
In this thesis, an extensive review for the previous work done in iris liveness detection is presented. The review has explored in detail the software-based and the hardware-based methods developed for iris liveness detection along with the different attacks detected and databases utilized for performance evaluation.
This thesis also explores the results of BSIF (Binarized statistical image features) descriptor for solving the problem of iris liveness detection to combat presentation attacks. A comprehensive study for the impact of segmentation on iris liveness detection has been carried out. Four public datasets representing printed, plastic, synthetic and contact lens attacks were used for method evaluation in both scenarios segmented and unsegmented eye images. The results have showed that BSIF can efficiently detect plastic and synthetic attacks without segmentation with correct classification rate of 100%. In addition, unsegmented eye images achieved better results in detecting print attack on the tested datasets. Segmentation is still required in the most challenging contact lens attacks.
A new method is proposed using residual images with BSIF to enhance the results of contact lens databases. Three high pass filters were applied separately before feature extraction with BSIF. The results were promising in the unsegmented scenario and the three filters enhanced in the results with 8.6667%, 10% and 18.3333%.
The main contribution is enhancing the accuracy of BSIF on Clarkson dataset from 91.67% to 93.33% in segmented mode using the first filter, in addition to proving that using the whole eye image is better in case of visible-light mobile iris datasets.
This thesis is a comprehensive study that try to evaluate the segmentation step value in iris liveness detection using BSIF descriptor.