Logo des Repositoriums

Auflistung nach Autor:in "Deravi, Farzin"

1 - 4 von 4
  • Konferenzbeitrag
    Performance Evaluation of Multibiometric Face Recognition Systems
    (BIOSIG 2008: Biometrics and Electronic Signatures, 2008) Castro Neves, Margarida; Chindaro, Samuel; Ng, Ming; Zhou, Ziheng; Deravi, Farzin
    The 3D Face project investigates the use of 3D face recognition technologies, and aims to improve their performance so that it would be possible to use such technologies in unsupervised access control scenarios in airports. During this project novel sensors, 2D, 3D, skin texture matchers and fusion algorithms have been developed. A technology performance test has been performed on all algorithms, in order to evaluate the technology improvements. This paper describes the independent test and evaluation activities for this project and gives an overview of the results obtained.
  • Konferenzbeitrag
    Quality filtering of EEG signals for enhanced biometric recognition
    (BIOSIG 2013, 2013) Yang, Su; Deravi, Farzin
    In this paper we present a biometric person recognition system based on EEG signals incorporating a novel strategy to find and utilize the most informative data segments using the concept of Sample Entropy. The users are presented with a stimulus that prompts a motor-imagery response. This is then measured using an array of EEG sensors. A sliding-window segmentation scheme and Wavelet Packet Decomposition are adopted for primary feature extraction before the quality measurement stage. The quality-filtered feature windows are then used to extract secondary features that are in turn classified using a linear discriminant classifier. The proposed system is tested using a publicly available EEG database and it shows that entropy filtering results in a significant improvement on performance. An average identification accuracy rate of more than 90% is achieved for 109 subjects using only eight electrodes, utilizing only the highest quality for each subject
  • Konferenzbeitrag
    Spoofing attempt detection using gaze colocation
    (BIOSIG 2013, 2013) Ali, Asad; Deravi, Farzin; Hoque, Sanaul
    Spoofing attacks on biometric systems are one of the major impediments to their use for secure unattended applications.This paper presents a novel method for face liveness detection by tracking the gaze of the user with an ordinary webcam. In the proposed system, an object appears randomly on the display screen which the user is required to look at while their gaze is measured. The visual stimulus appears in such a way that it repeatedly directs the gaze of the user to specific points on the screen. Features extracted from images captured at these sets of colocated points are used to estimate the liveness of the user. A scenario is investigated where genuine users track the challenge with head/eye movements whereas the impostors hold a photograph of the target user and attempt to follow the stimulus during simulated spoofing attacks. The results from the experiments indicate the effectiveness of the gaze colocation feature in detecting spoofing attack.
  • Konferenzbeitrag
    A versatile iris segmentation algorithm
    (BIOSIG 2011 – Proceedings of the Biometrics Special Interest Group, 2011) Radu, Petru; Sirlantzis, Konstantinos; Howells, Gareth; Hoque, Sanaul; Deravi, Farzin
    In biometric authentication, iris recognition was shown to be one of the most accurate techniques. In unconstrained environment and capture conditions (e.g. with hand-held devices used outdoors), the iris image may be contaminated by noise and distortions which makes the iris segmentation and recognition process difficult. This paper presents a novel iris segmentation algorithm that addresses some of the issues raised by unconstrained iris recognition. There are two main contributions in the present work: first, the proposed segmentation algorithm is able to cope efficiently with both near infra red and visible spectrum images; second, the algorithm speed can be increased significantly with a minimal reduction in accuracy. The versatility of the algorithm has been tested using both near infrared iris images acquired with a hand-held device and colour iris images acquired in unconstrained environment.