BTSA 2013 Abstracts


Full Papers
Paper Nr: 2
Title:

Improving Video-based Iris Recognition Via Local Quality Weighted Super Resolution

Authors:

Nadia Othman, Nesma Houmani and Bernadette Dorizzi

Abstract: In this paper we address the problem of iris recognition at a distance and on the move. We introduce two novel quality measures, one computed Globally (GQ) and the other Locally (LQ), for fusing at the pixel level the frames (after a bilinear interpolation step) extracted from the video of a given person. These measures derive from a local GMM probabilistic characterization of good quality iris texture. Experiments performed on the MBGC portal database show a superiority of our approach compared to score-based or average image-based fusion methods. Moreover, we show that the LQ-based fusion outperforms the GQ-based fusion with a relative improvement of 4.79% at the Equal Error Rate functioning point.

Paper Nr: 6
Title:

Distance-based Algorithm for Biometric Applications in Meanwaves of Subject’s Heartbeats

Authors:

Tiago Araújo, Neuza Nunes, Hugo Gamboa and Ana Fred

Abstract: The authors present a new biometric classification procedure based on meanwave’s distances of electrocardiogram (ECG) heartbeats. The ECG data was collected from 63 subjects during two data-recording sessions separated by six months (Time Instance 1, T1, and Time Instance 2, T2). Two classification tests were performed with the goal of subject identification using a distance-based method with the heartbeat waves. In both tests, the enrollment template was composed by the averaging of the T1 waves for each subject. For the first test, we composed five meanwaves of different T1 waves; In the second test, five meanwaves of different groups of T2 waves were composed. Classification was performed through the implementation of a kNN classifier, using the meanwave’s Euclidean distances as features for subject identification. In the first test, with only T1 waves, 95.2% of accuracy was achieved. In the second test, using T2 waves to compose the dataset for testing, the accuracy was 90.5%. The T2 waves belonged to the same subjects but were acquired in different time instances, simulating a real biometric identification problem. We therefore conclude that a distance-based method using meanwaves of ECG heartbeats for each subject is a valid parameter for classification in biometric applications.

Short Papers
Paper Nr: 4
Title:

Keystroke Authentication on Mobile Devices with a Capacitive Display

Authors:

Matthias Trojahn and Frank Ortmeier

Abstract: Nowadays, it is common to address security problems in the newspaper. Losing or stealing of mobile devices (smartphones and tablets) is in particular an important topic. A lot of information can be stored and accessed via these devices. The one reason why this problem exists, is because the mobile devices are not secured properly. In our work we present an authentication method for these mobile devices.We are using the keystroke dynamics during typing a PIN (four or six numbers) or password. With this the security of the devices gets improved. Keystroke dynamics are already used for authentication on PCs and on mobile phones with hardware keys on a 12-key layout.

Paper Nr: 5
Title:

Optimal Bayes Classification of High Dimensional Data in Face Recognition

Authors:

Wissal Drira and Faouzi Ghorbel

Abstract: In the supervised context, we intend to introduce a system which is composed of a series of novel and efficient algorithms that is able to realize a non parametric Bayesian classifier for high dimension. The proposed system tries to search for the best discriminate sub space in the mean of the minimum of the probability error of classification which is computed by using a modified kernel estimate of the conditional probability density functions. Therefore, Bayesian classification rule is applied in the reduced sub space. Such heuristic consists of four tasks. First, we maximize a novel estimate of the quadratic measure of the probabilistic dependence in order to realize multivariate extractors resulting from a number of different initializations of a given numerical optimizing procedure. Second, an estimation of the miss classification error is computed for each solution by the kernel estimate of the conditional probability density functions with the optimal band-with parameter in the sense of the Mean Integrate Square Error (MISE) which is obtained with the Plug in algorithm. Third, the sub space which presents the minimum of the miss classification values is thus chosen. After that, the Bayesian classification rule is operated in the reduced sub space with the optimal MISE of the modified kernel estimate. Finally, different algorithms will be applied to a base of images in grayscale representing classes of faces, showing its interest in the case of real data.

Paper Nr: 7
Title:

An Overview on Multi-biometric Score-level Fusion - Verification and Identification

Authors:

Naser Damer, Alexander Opel and Andreas Shahverdyan

Abstract: Multi-biometrics is the use of multiple biometric recognition sources to provide a more dependable verification or identification decision. Fusion of multi-biometric sources can be performed on different levels, such as the data, feature, or score level. This work presents an overview of the multi-biometric score-level fusion problem, along with the proposed solution in the literature. A discussion is made to provide a comparison between multi-biometric fusion in both scenarios. This discussion aims at providing a clearer view of future developments especially under the identification scenario where many related applications are rapidly growing such as forensics and ubiquitous surveillance.