Iris presentation attack detection based on best- k feature selection from YOLO inspired RoI
- PDF / 3,156,992 Bytes
- 21 Pages / 595.276 x 790.866 pts Page_size
- 2 Downloads / 197 Views
(0123456789().,-volV)(0123456789(). ,- volV)
ORIGINAL ARTICLE
Iris presentation attack detection based on best-k feature selection from YOLO inspired RoI Meenakshi Choudhary1 • Vivek Tiwari1
•
Venkanna Uduthalapally1
Received: 16 November 2019 / Accepted: 3 September 2020 Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Obfuscating an iris recognition system through forged iris samples has been a major security threat in iris-based authentication. Therefore, a detection mechanism is essential that may explicitly discriminate between the live iris and forged (attack) patterns. The majority of existing methods analyze the eye image as a whole to find discriminatory features for fake and real iris. However, many attacks do not alter the entire eye image, instead merely the iris region is affected. It infers that the iris embodies the region of interest (RoI) for an exhaustive search towards identifying forged iris patterns. This paper introduces a novel framework that locates RoI using the YOLO approach and performs selective image enhancement to enrich the core textural details. The YOLO approach tightly bounds the iris region without any pattern loss, where the textural analysis through local and global descriptors is expected to be efficacious. Afterward, various handcrafted and CNN based methods are employed to extract the discriminative textural features from the RoI. Later, the bestk features are identified through the Friedman test as the optimal feature set and combined using score-level fusion. Further, the proposed approach is assessed on six different iris databases using predefined intra-dataset, cross-dataset, and combined-dataset validation protocols. The experimental outcomes exhibit that the proposed method results in significant error reduction with the state of the arts. Keywords DarkNet-19 Feature selection Image enhancement Iris presentation attack detection RoI localization Score-level fusion
1 Introduction Iris recognition (IR) has achieved vigorous research interest due to its peerless individualities such as the rich morphological structure, certain distinctiveness for individuals (even twins), and constancy in micro-features regardless of the growing age [1]. Nevertheless, the IR systems are susceptible to presentation attacks that attempt to emasculate the application security. These attacks represent the forged or deliberately designed iris patterns in
& Vivek Tiwari [email protected] Meenakshi Choudhary [email protected] Venkanna Uduthalapally [email protected] 1
DSPM IIIT Naya Raipur, Naya Raipur, CG 493661, India
front of the iris camera/sensor to obstruct the functioning of the IR system [2]. These may be used to register contrived irises, purposely obscure a party’s trait, or even forge the iris pattern of another person [3]. There are several ways to reproduce the iris patterns, such as using textured contact lenses, printed iris images, artificial eyeballs, and playing iris images/videos on the LCD, and drug-prompted iris employmen
Data Loading...
