Publications
For a short comprehensive list of my publications, please check my Curriculum vitae.
Journal articles
- Andreas Uhl, Peter Wild,
"Single-sensor multi-instance fingerprint and eigenfinger recognition using (weighted) score combination methods",
International Journal on Biometrics (Special Issue on Multimodal Biometric and Biometric Fusion) 1:4, pp. 442-462, 2009
Abstract
When multiple instances of single biometrics can be acquired from a single input simultaneously, a multiple-step acquisition at additional transaction time cost can be avoided. We present a rotation-invariant, peg-free multi-instance Fingerprint and Eigenfinger-based biometric system extracting multiple features from a palmar scan of the hand. Our evaluation targets (a) rankings of individual fingers with respect to Minutiae and Eigenfinger features; (b) fusion of multi- instance intra-feature (Minutiae or Eigenfinger) matching scores; (c) cross-feature compared to intra-feature performance; (d) optimal weights for weighted versions of five score-level fusion methods Max, Median, Min, Product and Sum, and (e) aspects of computational demands for hand-based identification discussing the usage of serial classifier combinations instead of classically employed parallel ones. We examine results of an experimental approach to the problem of finding a suitable fusion method by investigating the effect of matcher-specific combination weights on recognition accuracy and compare cross-feature and intra-feature score combinations. - Andreas Uhl, Peter Wild,
"Footprint-based biometric verification",
Journal of Electronic Imaging 17, pp. 011016, doi:10.1117/1.2892674, 2008
Abstract
We investigate the potential of foot biometric features based on geometry, shape and texture and present algorithms for a prototype rotation invariant verification system. An introduction to origins and fields of application for footprint-based personal recognition is accompanied by a comparison with traditional hand biometry systems. Image enhancement and feature extraction steps emphasizing on specific characteristics of foot geometry and their permanence and distinctiveness properties, respectively, are discussed. Collectability and universality issues are considered as well. A visualization of various test results comparing discriminative power of foot shape and texture is given. The impact on real-world scenarios is pointed out, and a summary of results is presented. - Michael Gschwandtner, Andreas Uhl, Peter Wild,
"Transmission error and compression robustness of 2D Chaotic Map image encryption schemes",
EURASIP Journal on Information Security 2007:Article ID 48179, pp. doi:10.1155/2007/48179, 16 pages, 2007
Abstract
This paper analyzes the robustness properties of 2D chaotic map image encryption schemes. We investigate the behavior of such block ciphers under different channel error types and find the transmission error robustness to be highly dependent on the type of error occurring and to be very different as compared to the effects when using traditional block ciphers like AES. Additionally, chaotic-mixing-based encryption schemes are shown to be robust to lossy compression as long as the security requirements are not too high. This property facilitates the application of these ciphers in scenarios where lossy compression is applied to encrypted material, which is impossible in case traditional ciphers should be employed. If high security is required chaotic mixing loses its robustness to transmission errors and compression, still the lower computational demand may be an argument in favor of chaotic mixing as compared to traditional ciphers when visual data is to be encrypted.
Conference papers
- Andreas Uhl, Peter Wild,
"Enhancing Iris Matching Using Levenshtein Distance with Alignment Constraints",
In G. Bebis, R. Boyle, B. Parvin, D. Koracin, R. Chung and R. Hammoud, editors, Advances in Visual Computing: 6th International Symposium, (ISVC 2010), pp. 469-479, LNCS, 6453, Springer Verlag, 2010
Abstract
Iris recognition from surveillance-type imagery is an active research topic in biometrics. However, iris identification in unconstrained conditions raises many proplems related to localization and alignment, and typically leads to degraded recognition rates. While development has mainly focused on more robust preprocessing, this work highlights the possibility to account for distortions at matching stage. We propose a constrained version of the Levenshtein Distance (LD) for matching of binary iris-codes as an alternative to the widely accepted Hamming Distance (HD) to account for iris texture distortions by e.g. segmentation errors or pupil dilation. Constrained LD will be shown to outperform HD-based matching on CASIA (third version) and ICE (2005 edition) datasets. By introducing LD alignment constraints, the matching problem can be solved in O(n*s) time and O(n+s) space with n and s being the number of bits and shifts, respectively. - Christian Rathgeb, Andreas Uhl, Peter Wild,
"Incremental Iris Recognition: A Single-algorithm Serial Fusion Strategy to Optimize Time Complexity",
In Proceedings of the 4th IEEE International Conference on Biometrics: Theory, Application, and Systems 2010 (IEEE BTAS'10), pp. 1-6, IEEE Press, Sept. 2010
Abstract
Daugman's algorithm, mapping iris images to binary codes and estimating similarity between codes applying the fractional Hamming Distance, forms the basis of today's commercially used iris recognition systems. However, when applied to large-scale databases, the linear matching of a single extracted iris-code against a gallery of templates is very time consuming and a bottleneck of current implementations. As an alternative to pre-screening techniques, our work is the first to present an incremental approach to iris recognition. We combine concentration of information in the first bits of an iris-code with early rejection of unlikely matches during matching stage to incrementally determine the best-matching candidate in the gallery. Our approach can transparently be applied to any iris-code based system and is able to reduce bit comparisons significantly (to about 5% of iris-code bits) while exhibiting a Rank-1 Recognition Rate being at least as high as for matches involving all bits. - Mario Konrad, Herbert Stögner, Andreas Uhl, Peter Wild,
"Computationally efficient serial combination of rotation-invariant and rotation compensating iris recognition algorithms",
In P. Richard, J. Braz, editors, Proceedings of the 5th International Conference on Computer Vision Theory and Applications, VISAPP'10, pp. 85-90, Angers, France, 1, May 17 - 21, 2010
Abstract
Rotation compensation is one of the computational bottlenecks in large scale iris-based identification schemes, since a significant amount of Hamming distance computations is required in a single match due to the necessary shifting of the iris codes to compensate for eye tilt. To cope with this problem, a serial classifier combination approach is proposed for iris-based identification, combining rotation-invariant pre-selection with a traditional rotation compensating iris code-based scheme. The primary aim, a reduction of computational complexity, can easily be met - at comparable recognition accuracy, the computational effort required is reduced to 20% or even less of the fully fledged iris code based scheme. As a by-product, the recognition accuracy is shown to be additionally improved in open-set scenarios. - Andreas Uhl, Peter Wild,
"Parallel Versus Serial Classifier Combination for Multibiometric Hand-based Identification",
In M. Tistarelli, M.S. Nixon, editors, Proceedings of the 3rd International Conference on Biometrics 2009 (ICB'09), pp. 950-959, LNCS, 5558, Springer Verlag, 2009
Abstract
This paper presents an approach for optimizing both recognition and processing performance of a biometric system in identification mode. Multibiometric techniques facilitate bridging the gap between desired performance and current unimodal recognition rates. However, traditional parallel classifier combination techniques, such as Score sum, Borda count and Highest rank, cause further processing overhead, as they require a matching of the extracted sample with each template of the system for each feature. We examine a framework of serial combination techniques, which exploits ranking capabilities of individual features by reducing the set of possible matching candidates at each iteration, and we compare its performance with parallel schemes. Using this technique, both a reduction of misclassification and processing time in identification mode will be shown to be feasible for a single-sensor hand-based biometric system. - Andreas Uhl, Peter Wild,
"Comparing Verification Performance of Kids and Adults for Fingerprint, Palmprint, Hand-geometry and Digitprint Biometrics",
In Proceedings of the 3rd IEEE International Conference on Biometrics: Theory, Application, and Systems 2009 (IEEE BTAS'09), pp. 1-6, IEEE Press, Oct. 2009
Abstract
With the large scale deployment of biometrics for access control in private and public places, systems are faced the challenge of processing a diverse range of people. Most systems have been well evaluated for adults, however, their application in schools, or for private door access control, raises the question, whether there exists significant difference in performance between age groups in general and between kids and adults in particular. This paper targets an evaluation of the impact of children as biometric users on recognition accuracy for a series of hand-based modalities: Fingerprint, Palmprint, Hand-geometry and Digitprint. Furthermore, we try to analyze reasons for child-aging effects on performance at both feature and instance level using our database of 301 kids and 86 adults. - Andreas Uhl, Peter Wild,
"Personal Recognition using single-sensor multimodal hand biometrics",
In A. Elmoataz, O. Lezoray, F. Nouboud, D. Mammass, editors, Image and Signal Processing. Proceedings of ICISP 2008, pp. 396-404, LNCS, 5099, Springer Verlag, 2008
Abstract
Single-sensor approaches to multimodal biometric authentication targeting the human hand in multiple-matcher scenarios provide higher security in terms of accuracy and resistance to biometric system attacks than unimodal systems. This paper introduces a novel multimodal hand biometric system using palmar images acquired by a commercially available flatbed scanner. Hence, the presented approach to personal recognition is independent of specific biometric sensors, such as fingerprint readers or palmprint scanners. Experimental results with a minimum half total error rate of 0.003% using a database of 443 hand images will illustrate the performance improvement when hand-geometry, fingerprint and palmprint-based features are combined. - Andreas Uhl, Peter Wild,
"Personal identification using Eigenfeet, Ballprint and Foot geometry biometrics",
In Proceedings of the First IEEE International Conference on Biometrics: Theory, Application, and Systems 2007 (IEEE BTAS'07), pp. 1-6, IEEE Press, Oct. 2007
Abstract
This paper presents a new approach for personal identification using foot-biometric features based on Eigenfeet, local ridge characteristics and shape geometry. Due to the immutability of personal biometric features questions regarding privacy issues arise, when e.g. fingerprints are compromised. For niche applications, employing footprint-based authentication might bear enough security without the drawback of relying on sensitive data demanded by high-security applications. Part one of this paper outlines origins, fields of application and proposed system setup. After an introduction to employed normalization and feature extraction steps we discuss specific characteristics of foot biometrics compared to traditional hand and face-based techniques. Experimental results of all three employed biometric matchers are given and analyzed in part two. Finally, we present a summary of observed results. - Michael Gschwandtner, Andreas Uhl, Peter Wild,
"Compression of encrypted Visual Data",
In H. Leitold, E. Markatos, editors, Communications and Multimedia Security, Proceedings of the 10th IFIP International CMS 2006 Conference, pp. 141-150, Lecture Notes on Computer Science, 4237, Springer Verlag, Oct. 2006
Abstract
Chaotic mixing based encryption schemes for visual data are shown to be robust to lossy compression as long as the security requirements are not too high. This property facilitates the application of these ciphers in scenarios where lossy compression is applied to encrypted material - which is impossible in case traditional ciphers should be employed. If high security is required chaotic mixing loses its robustness to compression, still the lower computational demand may be an argument in favor of chaotic mixing as compared to traditional ciphers when visual data is to be encrypted.
Theses
- Peter Wild,
"Single-sensor hand and footprint-based multimodal biometric recognition",
Master's Thesis, University of Salzburg, Austria, January 2008.
Abstract
Biometric systems support the task of reliable automatic authentication, which is a key function for economic transactions in modern society. So far, no universal biometric modality suitable for all applications has been found. This thesis examines the multimodal fusion of different modalities using a single high-resolution scan of the human hand as input and relates existing techniques to a new biometric modality: the human foot. With the target application of wellness areas and spas, this new modality supports privacy interests and still provides satisfying accuracy. After an introduction to basic design principles, related work in palmprint, fingerprint, hand geometry, and footprint-based recognition is discussed. System modules for sensing, preprocessing, feature extraction, matching and decision for both implemented prototype footprint and hand-based biometric systems are described in detail. Necessary modifications due to anatomical differences are proposed and presented. Finally, a performance evaluation comparing overall accuracy and relative performance of individual features concludes this work.