Priv.-Doz. Dr. Bernhard A. Moser

Scientific Head Knowledge Based Vision Systems
Phone: +43 7236 3343 833
Fax: +43 7236 3343 888


Data Science - KVS

Key researcher for (dis)similarity measures, metric analysis, quasi-isometry, discrete geometry, image and signal processing, and machine learning at the Software Competence Center Hagenberg, Austria (SCCH). SCCH is a research and technology organization funded within the Austrian COMET program for Competence Centers of Excellent Technologies. Main research interests are devoted to mathematical aspects of pattern recognition, signal processing and deep learning with focus on non-standard similarity measures, geometric approaches as well as probability-based, kernel- and rule-based methods.

Scope of scientific work encompasses

  • Program Committee Member of 9th International Joint Conference on Computational Intelligence, IJCCI 2017
  • Program Committee Member and Special Session Organizer at EBCCSP 2015, 2016, 2017 (Int. Conf. on Event-Based Control, Communication and Signal Processing)
  • Program Chair of the Austrian Association for Pattern Recognition 2016, OAGM/AAPR.
  • Coordination of the Austrian IKTdZ project “Hyperion3D” on 3D film production (2013-2015).
  • Initiation and coordination of a fundamental research project funded by the Austrian Science Fund on Hermann Weyl’s discrepancy measure as image similarity (FWF P21496, 2010-2013).
  • Initiation and coordination of the European eraSME project “texQuality3D“on “High-performance quality inspection of industrial fabrics by 3D profile analysis” (June 2010 – June 2012).
  • Publications in top ranked journals, e.g., IEEE-TSP (Trans. on  Signal Processing),  IEEE-TPAMI (Trans. on Pattern Analysis and Machine Intelligence), JLMR (Journal for Machine Learning Research), DCG (Discrete & Computational Geometry), EJC (Electronic Journal of Combinatorics)
  • Invited plenary talk at the 8th International Conference on Fuzzy Set Theory and Applications (FSTA) on “Kernels from a fuzzy perspective”, Liptovsky Jan, Slovakia, 2006.

Study and Employment

2017   Habilitation for Mathematics at JKU Linz
2013 now Scientific Head of Knowledge-Based Vision Systems and Research Coordinator at SCCH
2006 2013 Area Manager of Knowledge-Based Technology Area
2005 2006 Head of Computer Vision Group at SCCH
2001 2004 Cyberhouse GmbH / Software Engineering and Project Managment
1999 2001 PC Technik GmbH / Software Engineering and Project Managment
1994 1999 University of Linz / Research Assistant / Mathematics
1992 1993 University of Salzburg / Research Assistant / Computer Science
1991 1996 Doctoral study at Universities of Salzburg and Linz/ Fuzzy Logic
1986 1991 Study of Mathematics/University of Salzburg

Education

2005 present

Series of courses in project management (both classical approach and agile project management), communication, moderation, and marketing training;

1991 1996 Doctoral study in mathematics at Paris Lodron University of Salzburg and Johannes Kepler University Linz; thesis title “A New Approach for Representing Control Surfaces by Fuzzy Rule Bases”; supervisor Prof. Erich Peter Klement (University of Linz), co-supervisor Prof. Peter Zinterhof (University of Salzburg); graduation (“Dr. rer. nat.”) in February 1996 with first-class honors
1986 1991 Study of mathematics at Paris Lodron University of Salzburg; thesis in the field of mathematical logics, thesis title “Diodorean Modality of the Minkowsky space-time”; supervisor Prof. Johannes Czermak; graduation (“Mag. rer. nat.'') in October 1991 with first-class honors

Employment Record (last 10 years)

July 2006 present

Scientific Head for Knowledge-Based Vision Systems (about 10 people); Research Coordinator for SCCH with focus on stimulating the interdisciplinary research at SCCH.

July 2006 June 2013 Manager of the Knowledge-Based Technology area (about 25 people) at SCCH which covers the core competences of industrial data mining, knowledge-based vision systems for industrial applications and biomedical data analysis. Range of tasks comprises business development, positioning of this group, networking with other national and international research institutions and industrial partners, research and management of research projects (both academic and with industrial partners).
January 2005 July 2006  Head of Computer Vision Group at the Software Competence Center Hagenberg, SCCH. Research, development and project management of computer vision projects for industrial applications ranging from quality inspection systems, object detection and tracking for multi-media applications, multi-camera and 3D television.

Patents

  • German Patent, DE 10 2018 122 019.7, DPMA S2950-DE, on a method to verify the classifier integrity of deep neural networks, "Integrität neuronaler Netze", submitted 10th of Sept., 2018, B. Moser (inventor).
  • ANR 1415 / 2008; IPC G01N, patent applied for detecting structural defects in patterns with high regularities like woven fabrics, P. Haslinger and B. Moser (inventors), 2008, accepted 2010.
  • ANR 1131 / 2008; IPC G01B, patent applied for an illumination technique for detecting defects on reflecting surfaces Wieser and B. Moser (inventors), 2008.

Selection of Reviewing 

IEEE Transactions on Signal Processing, IEEE Transactions on Medical Imaging, IEEE Transactions on Fuzzy Systems, Journal of Electronic Imaging, Journal of Fuzzy Sets and Systems, Journal for Soft Computing, Lecture Notes in Computer Science (LNCS), Journal of Mathematical and Computer Modelling, Journal of Applied Mathematics Letters, Journal of Information Science, Mathematical Reviews; Information Sciences.

Journal Publications

  • H. Eghbal-zadeh, L. Fischer, N. Popitsch, F. Kromp, S. Taschner-Mandl, K. Koutini, T. Gerber, E. Bozsaky, P. F. Ambros, I. M. Ambros, G. Widmer, B. A. Moser, Deep SNP: An End-to-end Deep Neural Network with Attention-based Localization for Break-point Detection in SNP Array Genomic Data, Journal of Computational Biology (to be published), October 2018.
  • H. Eghbal-zadeh, L. Fischer, N. Popitsch, F. Kromp, S. Taschner-Mandl, K. Koutini, T. Gerber, E. Bozsaky, P. F. Ambros, I. M. Ambros, G. Widmer, B. A. Moser, “Deep SNP: An End-to-end Deep Neural Network with Attention-based Localization for Break-point Detection in SNP Array Genomic data”, Joint ICML and IJCAI 2018 Workshop on Computational Biology, June 2018 (to appear in Journal of Computational Biology)
  • B. A. Moser, Similarity Recovery from Threshold-Based Sampling under General Conditions, IEEE Transactions on Signal Processing, Vol. 65, Issue 17, 4645 – 4654, 2017.
  • W. Reisner, B. Moser, Improving Visual Discomfort Prediction for Stereoscopic Images via Disparity-Based Contrast, Journal of Imaging Science and Technology, 11/2015; 59(6):60401-1-60401-8. DOI: 10.2352/J.ImagingSci.Technol.2015.59.6.060401.
  • R. Leser, B. Moser, T. Hoch, J. Stöger, G. Kellermayr, S. Reinsch, A. Baca. Expert-oriented modelling of a 1vs1-situation in football. International Journal of Performance Analysis in Sport, volume 15, number 3, pages 949-966, December, 2015.
  • B. Moser. The Range of a Simple Random Walk on Z: An Elementary Combinatorial Approach, Electronic Journal of Combinatorics, 21(4):#P4.10, 2014.
  • B. Moser. On a Multisection Style Binomial Summation Identity for Fibonacci Numbers, Int. J. Contemp. Math. Sciences, 9(4):175--186, 2014.
  • B. Moser and T. Natschläger. On Stability of Distance Measures for Event Sequences Induced by Level-Crossing Sampling, IEEE Trans. on Signal Processing, 62(8):1987--1999, 2014.
  • B. Moser: Geometric characterization of Weyl’s Discrepancy Norm in terms of its n-dimensional unit balls. Journal Discrete & Computational Geometry (DCG), volume 48, number 4, pages 793-806, 2012.
  • H. Schöner, F. Bauer, A. Dorrington, B. Heise, V. Wieser, A. Payne, M. Cree, B. Moser. Image processing for 3D-scans generated by time of flight range cameras. Journal of Electronic Imaging, volume 21, number 2, DOI:10.1117/1.JEI.21.2.023012, 2012.
  • V. Wieser, C. Grelck, J. Guo, P. Haslinger, F. Korzeniowski, B. Moser, S. Scholz. Combining high productivity and high performance in image processing using single assignment C on multi-core CPUs and many-core GPUs. Journal of Electronic Imaging, volume 21, number 2, DOI: 10.1117/1.JEI.21.2.021116, 2012.
  • B. Moser. A similarity measure for image and volumetric data based on Hermann Weyl's discrepancy. IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 33, number 11, pages 2321-2329, 2011.
  • B. Moser: On the compactness of admissible transformations of fuzzy partitions in terms of T-equivalence relations. Fuzzy Sets and Systems, vol. 160 (1), pp. 95-106, 2009.
  • B. Moser: On the T-Transitivity of Kernels. Fuzzy Sets and Systems, vol 157, pp. 1787-1796, 2006.
  • B. Moser: On Representing and Generating Kernels by Fuzzy Equivalence Relations. Journal of Machine Learning Research, MIT Press Cambridge, MA, USA, vol 7, 2603-2620, 2006.
  • B. Moser and M. Navara. Fuzzy Controllers with Conditionally Firing Rules. IEEE Trans. Fuzzy Systems, 10(3):340–348, 2002.
  • D. Tikk, I. Joó, L. T. Kóczy, P. Várlaki, B. Moser, and T. D. Gedeon. Stability of interpolative fuzzy KH-controllers. Fuzzy Sets and Systems, 125:105–119, 2002.
  • B. Moser and R. Winkler. A relationship between equality relations and the T-redundancy of fuzzy partitions and its application to Sugeno controllers. Fuzzy Sets and Systems, 114:455–467, 2000.
  • E. P. Klement, L. T. Kóczy and B. Moser. Are Fuzzy Systems Universal Approximators? Int. J. General Systems, 28(2–3):259–282, 1999.
  • B. Moser. Sugeno controllers with a bounded number of rules are nowhere dense. Fuzzy Sets and Systems, 104(2):269–277, 1999.
  • B. Moser, E. Tsiporkova and E. P. Klement. Convex combinations in terms of triangular norms: A characterization of idempotent, bisymmetrical and self-dual compensatory operators. Fuzzy Sets and Systems, 104:97–108, 1999.
  • E. P. Klement and B. Moser. On the T-redundancy of fuzzy partitions. Fuzzy Sets and Systems, 85:195–201, 1997.

Papers in Proceedings of Conferences with Peer Review

  • W. Zellinger, B. Moser, H. Eghbal-zadeh, M. Zwick, E. Lughofer, T. Natschlaeger, S. Saminger-Platz,“ Moment distances for comparing high-entropy distributions with application in domain adaptation“, 23rd International Conference on Computational Statistics (COMPSTAT 2018), Iasi, Romania, 28-31 August 2018.
  • H. Eghbal-zadeh, L. Fischer, N. Popitsch, F. Kromp, S. Taschner-Mandl, K. Koutini, T. Gerber, E. Bozsaky, P. F. Ambros, I. M. Ambros, G. Widmer, B. A. Moser, Deep SNP: An End-to-end Deep Neural Network with Attention-based Localization for Break-point Detection in SNP Array Genomic Data, arxiv.org/abs/1806.08840, accepted at the Joint ICML and IJCAI 2018 Workshop on Computational Biology, July 2018.
  • N. Shepeleva, T.Hoch, L. Fischer, W. Kloihofer, B.Moser, Removing nuisance in tracklet data, Deep Learning and Video Content Analysis, 10802-28, Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies, SPIE Security and Defense, 10-11 Sept. 2018.
  • H. Eghbal-zadeh, L. Fischer, N. Popitsch, F. Kromp, S. Taschner-Mandl, K. Koutini, T. Gerber, E. Bozsaky, P. F. Ambros, I. M. Ambros, G. Widmer, B. A. Moser, Deep SNP: An End-to-end Deep Neural Network with Attention-based Localization for Break-point Detection in SNP Array Genomic data, Joint ICML and IJCAI 2018 Workshop on Computational Biology. dx.doi.org/10.1101/354423, 2018.
  • B. A. Moser, On the Discrepancy Normed Space of Event Sequences for Threshold-based Sampling, EBCCSP 2018, arXiv:1806.06273v1.
  • W. Zellinger, B. Moser, A. Chouikhi, M. Gelautz, F. Seitner and M. Nezveda. Linear optimization approach for depth range adaption of stereoscopic videos, in IS&T Electronic Imaging: Stereoscopic Displays and Applications XXVII Proceedings, San Francisco, CA, USA, 15-17 Febr., 2016.
  • F. Seitner, M. Nezveda, M. Gelautz, G. Braun, C. Kapeller, W. Zellinger and B. Moser.
  • Trifocal system for high-quality inter-camera mapping and virtual view synthesis; International Conference on 3D Imaging (IC3D), Liège, Belgium; Dec 14-15, 2015, pages 1-8, DOI 10.1109/IC3D.2015.7391819.
  • B. Moser. Matching Event Sequences Approach based on Weyl's Discrepancy Norm, 1st. Int. Conf. on Event-based Control, Communication, and Signal Processing, EBCCSP 15, June 17-19, Krakow, Poland, 2015,  DOI 10.1109/EBCCSP.2015.7300676.  
  • B. Moser. Stability of Threshold-based Sampling as Metric Problem, 1st. Int. Conf. on Event-based Control, Communication, and Signal Processing, Special Session on Mathematical Modeling of Event-based Sampling, EBCCSP 15, June 17-19, Krakow, Poland, 2015, DOI 10.1109/EBCCSP.2015.7300692.
  • G. Stübl, P. Haslinger, V. Wieser, J. Scharinger, and B. Moser. Periodicity Estimation of Nearly Regular Textures based on Discrepancy Norm. In Proc. SPIE 8661, Image Processing: Machine Vision Applications VI, pp. 866106--866106-10. Burlingame, California, USA, 2013.
  • G. Stübl, J.-L. Bouchot, P. Haslinger, and B. Moser. Discrepancy Norm as Fitness Function for Defect Detection on Regularly Textured Surfaces. In A. Pinz, T. Pock, H. Bischof, and F. Leberl, editors, Pattern Recognition, volume 7476 of Lecture Notes in Computer Science, pp. 428--437, Springer Berlin Heidelberg, 2012.
  • B. Moser, G. Stübl, and J. L. Bouchot. On a Non-Monotonicity Effect of Similarity Measures.
  • In M. Pelillo and E.R. Hancock, editors, 1st Int. Conf. on Similarity-based Pattern Recognition, SIMBAD'11, vol. 7005 of Lecture Notes in Computer Science, pp. 46--60, Springer Berlin Heidelberg, 2011.
  • G. Stübl, B. Moser, J. Scharinger. On approximate nearest neighbor field algorithms in template matching for surface quality inspection In A. Quesada-Arencibia et al. (editors), Computer Aided Systems Theory - EUROCAST 2013,  Lecture Notes in Computer Science, vol. 8112,  pp 79-86, 2013.
  • V. Wieser, C. Grelck, H. Schöner, P. Haslinger, K. Bosa, B. Moser. GPU-based image processing use cases: A high-level approach. In K. De Bosschere, E.H. D'Hollander, G.R. Joubert, D. Padua, F. Peters (editors). Applications, Tools and Techniques on the Road to Exascale Computing, Advances in Parallel Computing, pages 199-206, IOS Press. 2012.
  • V. Wieser, B. Moser, S. Scholz, S. Herhut, J. Guo. Combining high productivity and high performance in image processing using single assignment C. In J.-C. Pinoli, J. Debayle, Y. Gavet, F. Gruy, C. Lambert (editors). Proc. SPIE 8000, 10th International Conference on Quality Control by Artificial Vision (QCAV'2011), DOI: 10.1117/12.890920, Society of Photo-Optical Instrumentation Engineers, 2012.
  • J. Bouchot, G. Stübl, B. Moser. A template matching approach based on the discrepancy norm for defect detection on regularly textured surfaces. In J. Pinoli, J. Debayle, Y. Gavet, F. Gruy, C. Lambert (editors), Proc. SPIE 8000, 10th International Conference on Quality Control by Artificial Vision (QCAV'2011), DOI: 10.1117/12.889865, Society of Photo-Optical Instrumentation Engineers, 2012.
  • J. Bouchot, J. Himmelbauer, and B. Moser. On autocorrelation based on Hermann Weyl's discrepancy norm for time series analysis. In Proc. 2010 Int. Joint Conf. on Neural Networks (IJCNN'10), pages 1-7, 2010.
  • C. Gonzalez-Mocillo, L. Lopez, J. Castro-Schez, B. Moser. A data-mining approach to 3D realistic render setup assistance. In K. Haigh, N. Rychtyckjy (editors). Proceedings of the 21st Innovative Applications of Artificial Intelligence Conference IAAI 2009, pages 93-98, AAAI Press. 2009.
  • H. Schöner, B. Moser, A. Dorrington, A. Payne, M. Cree, B. Heise, F. Bauer. A clustering based denoising technique for range images of time of flight cameras. In M. Masoud Mohammadian (editor). Proceedings of the International Conference on Computational Intelligence for Modelling, Control and Automation (CIMCA 2008), pages 989-994, IEEE Computer Society. 2008
  • H. Schöner, B. Moser, E. Lughofer. On preprocessing multi-channel sensor data for online process monitoring. In M. Masoud Mohammadian (editor). Proceedings of the International Conference on Computational Intelligence for Modelling, Control and Automation (CIMCA 2008), pages 414-419, IEEE Computer Society. 2008.
  • B. Moser, P. Haslinger, and T. Kazmar. On the potential of Hermann Weyl's discrepancy norm for texture analysis. In Proc. 2008 Int. Conf. on Computational Intelligence for Modeling, Control and Automation; Intelligent Agents, Web Technologies and Internet Commerce; and Innovation in Software Engineering (CIMCA/IAWTIC/ISE), pages 187-191. IEEE Computer Society, 2008.
  • B. Moser and T. Hoch. Misalignment measure based on Hermann Weyl's discrepancy. In Proceedings of 32nd Workshop of the Austrian Association for Pattern Recognition (AAPT/OAGM), Challenges in the Biosciences: Image Analysis and Pattern Recognition Aspects, volume 232, pages 187-198, Linz, Austria, May 2008.
  • B. Moser and U. Bodenhofer. Correspondences between fuzzy equivalence relations and kernels: theoretical results and potential applications. In Proc. 15th IEEE Int. Conf. on Fuzzy Systems, pages 10217-10223, Vancouver, BC, July 2006.
  • B. Moser and M. Navara. Conditionally Firing Rules Extend the Possibilities of Fuzzy Controllers. In Proc. Int. Conf. Computational Intelligence for Modelling, Control and Automation, pp. 242-245, 1999.
  • B. Moser and M. Navara. Which triangular norms are convenient for fuzzy controllers. In Proc. EUSFLAT-ESTYLF Joint Conf. 99, Universitat de les Illes, pp. 75-78, 1999.

Contributions to Edited Volumes

  • V. Wieser, C. Grelck, H. Schöner, P. Haslinger, K. Bosa, B. Moser. GPU-based image processing use cases: A high-level approach. In K. De Bosschere, E.H. D'Hollander, G.R. Joubert, D. Padua, F. Peters (editors). Applications, Tools and Techniques on the Road to Exascale Computing, Advances in Parallel Computing, pages 199-206, IOS Press. 2012.
  • W. Beer, W. Kurschl, F.- Matusek, B. Moser, S. Mitsch, S. Sutor: Application Development and Management of Smart Camera Networks. Chapter 14 of “Smart Cameras”, A.N. Belbachier (ed.), Springer Science+Business Media, 2009, pp. 259-266, Springer US, 2009.
  • W. Schreiner (ed), K. Bosa, A. Langegger, T. Leitner, B. Moser, S. Pall, V. Wieser, W. Wöß:
    Parallel, Distributed, and Grid Computing. Chapter VII of "Hagenberg Research", B. Buchberger et al (eds), pp. 333-378, Springer, 2009.
  • E. P. Klement, E. Lughofer, J. Himmelbauer, B. Moser: Data-Driven and Knowledge-Based Modelling. Chapter of „Hagenberg Research“, B. Buchberger et al., Springer, pp. 237-279, Springer 2009.
  • P. Bauer, E. P. Klement, B. Moser, and A. Leikermoser. Modeling of Control Functions by Fuzzy Controllers. In H. T. Nguyen, M. Sugeno, R. M. Tong and R. R. Yager, editors, Theoretical Aspects of Fuzzy Control., chapter 5, 91–116. John Wiley & Sons, New York, 1995.
  • P. Bauer, E. P. Klement, A. Leikermoser and B. Moser, Interpolation and approximation of real input-output functions using fuzzy rule bases. In R. Kruse, J. Gebhardt, R. Palm, editors, Fuzzy Systems in Computer Science., 245–254, Vieweg, Braunschweig, 1994.

arXiv (not yet published)

W. Zellinger, B. A. Moser, T. Grubinger, E. Lughofer, T. Natschläger, S. Saminger-Platz, Robust Unsupervised Domain Adaptation for Neural Networks via Moment Alignment, arXiv:1711.06114.

Theses

  • B. Moser. Diodorean modality of the linear time and the Minkowsky space-time. Master’s Thesis, Paris Lodron University Salzburg, October 1991.
  • B. Moser. A new approach for representing control surfaces by fuzzy rule bases. PhD Thesis, Johannes Kepler University Linz, February 1996 (mentor Prof. E. P. Klement).

Membership of Commitees and Refereeing Activities

  • PC Member of EBCCSP and organisation of Special Session on “Mathematical Modeling of Event-based Sampling”, Krakow, Poland, 2015, 2016, Madeira 2017.
  • Co-Organisation of OAGM, Program Chair, Wels, Austria, 2016.
  • PC Member of QCAV, Lyon, France, 2011.
  • Member of advisory board for tech2B (Austrian institution for start-up companies, Linz), since 2008.
    PC Member of OAGM, Linz, 2008.