- KonferenzbeitragPredicting Dactyloscopic Examiner Fingerprint Image Quality Assessments(BIOSIG 2015, 2015) Olsen, Martin Aastrup; Böckeler, Martin; Busch, Christoph; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasWe work towards a system which can assist dactyloscopic examiners in assessing the quality and decision value of a fingerprint image and eventually a fingermark. However when quality assessment tasks of datyloscopic examiners are replaced by automatic quality assessment then we need to ensure that the automatic measurement is in agreement with the examiner opinion. Under the assumption of such agreement, we can predict the examiner opinion. We propose a method for determining the examiner agreement on ordinal scales and show that there is a high level of agreement between examiners assessing the ground truth quality of fingerprints. With ground truth quality information on 749 fingerprints and using 10-fold cross validation we construct models using Support Vector Machines and Proportional Odds Logistic Re- gression which predicts median examiner quality assessments 35\% better than when using the prior class distribution.
- KonferenzbeitragPrivacy preserving technique for set-based biometric authentication using Reed-Solomon decoding(BIOSIG 2015, 2015) Hartloff, Jesse; Mandal, Avradip; Roy, Arnab; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasIn this work, we present a single-factor biometric authentication system that provides template security against an adversarial server while allowing errortolerant matching. Our approach is to secure templates represented as sets using errorcorrecting codes and Reed-Solomon decoding. To accomplish this, each element in the set is combined with a random codeword and a secret share is computed using the codeword and a Reed-Solomon based secret sharing scheme. These random codewords provide uncertainty for an attacker, while the genuine user can decode to the correct values for verification. Without a reading from the enrolling biometric the shares will appear random, thus protecting the users biometric. We show implementation results for this system on fingerprints using pairs of minutia points. Our system overcomes many common weaknesses for template security systems including replay attacks, malicious servers, eavesdroppers, and record multiplicity attacks.
- KonferenzbeitragIdentification performance of evidential value estimation for fingermarks(BIOSIG 2015, 2015) Kotzerke, Johannes; Davis, Stephen A.; Hayes, Robert; Spreeuwers, Luuk J.; Veldhuis, Raymond N. J.; Horadam, Kathy J.; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasLaw enforcement agencies around the world use biometrics and fingerprints to solve and fight crime. Forensic experts are needed to record fingermarks at crime scenes and to ensure those captured are of evidential value. This process needs to be automated and streamlined as much as possible to improve efficiency and reduce workload. It has previously been demonstrated that is possible to estimate a fingermark's evidential value automatically for image captures taken with a mobile phone or other devices, such as a scanner or a high-quality camera. Here we study the relationship between a fingermark being of evidential value and its correct and certain identification and if it is possible to achieve identification despite the mark not having sufficient evidential value. Subsequently, we also investigate the influence the capture device used makes and if a mobile phone is an option worth considering. Our results show that automatic identification is possible for 126 of the 1 428 fin- , germarks captured by a mobile phone, of which 116 were marked as having evidential value by experts and 123 by an automated algorithm.
- KonferenzbeitragA signature complexity measure to select reference signatures for online signature verification(BIOSIG 2015, 2015) Kahindo, Christian; Garcia-Salicetti, Sonia; Houmani, Nesma; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasThis paper presents an original procedure for selecting the reference online signature instances of a writer, an important issue for any effective signature verifier. To this end, for each signature instance, we propose a novel complexity measure, by exploiting a global description of signatures in the frequency domain as well as a global statistical modelling of each signature instance. To select the reference signatures, we propose a method based on the distribution of complexity values for all the available genuine signatures. The 2500 genuine samples of MCYT-100 online database are used in this study. Experimental results show the effectiveness of the method and of the here proposed complexity measure for this specific task.
- KonferenzbeitragDiscarding low quality minutia cylinder-code pairs for improved fingerprint comparison(BIOSIG 2015, 2015) Izadi, M. Hamed; Drygajlo, Andrzej; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasLocal minutiae descriptors such as Minutia Cylinder-Code (MCC) are becoming increasingly popular in modern fingerprint verification systems. The verification performance depends on the fingerprint image quality in global and local levels. Discarding part of the lowest quality samples based on quality measures is a universal approach being widely used for improving the performance of biometric recognition systems. In this work, we evaluate several different discarding methods to filter out low quality pairs of MCC descriptors using minutiae qualities, with the final aim of improving global comparison accuracy. Moreover, we propose an efficient MCC based fingerprint comparison method based on discarding the low quality elements from local similarity matrix. Our extensive experiments on three different databases (FVC2002 DB2, FVC2002 DB3 and FVC2004 DB3) show that 1) the proper discarding of low quality MCC pairs from local similarity matrix either independently or using pairwise measures can improve the MCC based comparison performance, 2) for the proposed discarding method, the quality of central minutiae is more efficient as cylinder quality measure than the average minutiae qualities in each descriptor.
- Editiertes BuchBIOSIG 2015(2015) Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, Andreas
- KonferenzbeitragOn reducing the effect of silhouette quality on individual gait recognition: a feature fusion approach(BIOSIG 2015, 2015) Jia, Ning; Sanchez, Victor; Li, Chang-Tsun; Mansour, Hassan; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasThe quality of the extracted gait silhouettes can hinder the performance and practicability of gait recognition algorithms. In this paper, we propose a framework that integrates a feature fusion approach to improve recognition rate under this situation. Specifically, we first generate a dataset containing gait silhouettes with various qualities based on the CASIA Dataset B. We then fuse gallery data with different qualities and project data into embedded subspaces. We perform classification based on the Euclidean distances between fused gallery features and probe features. Experimental results show that the proposed framework can provide important improvements on recognition rate.
- KonferenzbeitragSegmentation-level fusion for iris recogntion(BIOSIG 2015, 2015) Wild, Peter; Hofbauer, Heinz; Ferryman, James; Uhl, Andreas; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasThis paper investigates the potential of fusion at normalisation/segmentation level prior to feature extraction. While there are several biometric fusion methods at data/feature level, score level and rank/decision level combining raw biometric signals, scores, or ranks/decisions, this type of fusion is still in its infancy. However, the increasing demand to allow for more relaxed and less invasive recording conditions, especially for on-the-move iris recognition, suggests to further investigate fusion at this very low level. This paper focuses on the approach of multi-segmentation fusion for iris biometric systems investigating the benefit of combining the segmentation result of multiple normalisation algorithms, using four methods from two different public iris toolkits (USIT, OSIRIS) on the public CASIA and IITD iris datasets. Evaluations based on recognition accuracy and ground truth segmentation data indicate high sensitivity with regards to the type of errors made by segmentation algorithms.
- KonferenzbeitragStructured forest edge detectors for improved eyelid and iris segmentation(BIOSIG 2015, 2015) Happold, Michael; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasEdge detection has long figured into iris segmentation algorithms, often providing a first-pass estimate of the inner and outer iris boundaries. Standard edge detectors, however, generally produce too many extraneous edges inside and outside the iris to be used for simple ellipse fitting to robustly find the iris boundaries. Solutions to this problem have been either to have additional filtering to select relevant edges or to design specialized edge detectors that highlight iris boundary edges and suppress irrelevant edge types. This description holds ever so more for eyelid boundaries, which are often very subtle. An edge detector that will trigger on an eyelid boundary will also likely trigger on almost any slight intensity gradient in an image. We seek to solve this problem by learning specialized edge detectors for each type of relevant boundary in an iris image. Using a fast Structured Random Forest approach developed for learning generalized edge detectors, we train detectors for the iris/sclera, iris/pupil, and eyelid boundaries. The results show that learned edge detectors should become part of the standard toolbox for iris segmentation and eyelid boundary detection.
- KonferenzbeitragLiveness Detectionin Biometrics(BIOSIG 2015, 2015) Krieg, Maximilian; Rogmann, Nils; Brömme, Arslan; Busch, Christoph; Rathgeb, Christian; Uhl, AndreasThe use of biometrics as an alternative for PIN and password-based authentication systems becomes increasingly attractive in this day and age. Biometric systems perform an authentication by personal biological or behavioral characteristics. A negative side effect of the increasing use of biometric systems is the progressing development of sophisticated attacks, in particular presentation attacks. Liveness detection has the aim to identify a living and during the biometric authentication process present individual as such and to repel spoofing attacks at the data capture subsystem. In this paper different current attack scenarios are described. Based on these scenarios, several liveness detection techniques are elaborated and investigated as possible countermeasures.