Duncan B.,University of Aberdeen |
Bratterud A.,University of Oslo |
Happe A.,Institute for Safety Technology GmbH
2016 6th International Conference on Innovative Computing Technology, INTECH 2016 | Year: 2016
Achieving cloud security is not a trivial problem and developing and enforcing good cloud security controls is a fundamental requirement if this is to succeed. The very nature of cloud computing can add additional problem layers for cloud security to an already complex problem area. We discuss why this is such an issue, consider what desirable characteristics should be aimed for and propose a novel means of effectively and efficiently achieving these goals through the use of unikernel based systems. The main thrust of this paper is to discuss the key issues which need to be addressed, noting which of those might be covered by our proposed approach. We discuss how our proposed approach may help better address the key security issues we have identified. © 2016 IEEE.
Kozlov M.,MR Computer GmbH |
Kozlov M.,Max Planck Institute for Human Cognitive and Brain Sciences |
Schaefers G.,MR Computer GmbH |
Schaefers G.,Institute for Safety Technology GmbH
Proceedings of the 12th IASTED International Conference on Biomedical Engineering, BioMed 2016 | Year: 2016
For insulated stainless steel wires of 1.5 mm in diameter with 0.5 mm insulation thickness, we evaluated the dependence of 64 MHz RF-induced power deposited at a hotspot (p) on: 1) lead length, 2) lead insulating electrical properties, 3) lead surrounding medium. Lead transfer functions (TF) were obtained with 3-D electromagnetic simulations. TF and p depended significantly on electrical properties of the insulation and lead surrounding medium. Increased insulator conductivity resulted in decreased p in most investigated cases. It is impossible to define one test surrounding medium that results in worst case power deposition for all cases: a) different lead lengths; b) different electrical properties of the lead insulator; and c) different tissues surrounding the lead in human body.
Huber-Mork R.,Institute for Safety Technology GmbH |
Zambanini S.,Vienna University of Technology |
Zaharieva M.,Vienna University of Technology |
Kampel M.,Vienna University of Technology
Machine Vision and Applications | Year: 2011
We present a vision-based approach to ancient coins' identification. The approach is a two-stage procedure. In the first stage an invariant shape description of the coin edge is computed and matching based on shape is performed. The second stage uses preselection by the first stage in order to refine the matching using local descriptors. Results for different descriptors and coin sides are combined using naive Bayesian fusion. Identification rates on a comprehensive data set of 2400 images of ancient coins are on the order of magnitude of 99%. © 2010 Springer-Verlag.
Papp A.,Institute for Safety Technology GmbH |
Wiesmeyr C.,Institute for Safety Technology GmbH |
Litzenberger M.,Institute for Safety Technology GmbH |
Garn H.,Institute for Safety Technology GmbH |
Kropatsch W.,Vienna University of Technology
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016
We propose a novel method for the detection of vibrations caused by trains in an optical fiber buried nearby the railway track. Using optical time-domain reflectometry vibrations in the ground caused by different sources can be detected with high accuracy in time and space. While several algorithms have been proposed in the literature for train tracking using OTDR signals they have not been tested on longer recordings. The presented method learns the characteristic pattern in the Fourier domain using a support vector machine (SVM) and it becomes more robust to any kind of noise and artifacts in the signal. The pointbased causal train tracking has two stages to minimize the influence of false classifications of the vibration detection. Our technical contribution is the evaluation of the presented algorithm based on two hour long recording and demonstration of open problems for commercial usage. © Springer International Publishing AG 2016.
Kraushaar U.,Multi Channel Systems |
Meyer T.,Reutlingen University |
Hess D.,Institute for Safety Technology GmbH |
Gepstein L.,Technion - Israel Institute of Technology |
And 3 more authors.
Expert Opinion on Drug Safety | Year: 2012
Introduction: The field of cardiac safety pharmacology has been experiencing exciting changes over the recent years. Drug induced arrhythmia of the torsade des pointes types has been the reason for the denial of approval of novel drug candidates. The aim of cardiac safety pharmacology is to detect undesirable pharmacodynamic drug effects within and above the therapeutic range. A special focus is on the identification of potential arrhythmogenic effects within the drug discovery chain. Areas covered: Here, the authors discuss the relevance of induced pluripotent stem (iPS) cell derived cardiomyocytes for safety pharmacology. The technology of obtaining functional cardiomyocytes from somatic cells of healthy donors and patients with inherited diseases is the basis for diverse disease models in multi-level safety pharmacology screening. The reader will gain an overview of stem cell based technologies in cardiac safety pharmacology in cardiac and disease modeling by iPS cell derived cardiomyocytes from patients with an inherited cardiac syndrome. Expert opinion: iPS cell derived cardiomyocytes especially from patients with increased risk of cardiac arrhythmia are on the verge of offering new options for drug testing. More reliable assays can be expected to predict the arrhythmogenic risk of drug candidates in humans. However, this technology is still new and extensive validation studies are due. © 2012 Informa UK, Ltd.
Schreier G.,Institute for Safety Technology GmbH |
Schreier G.,University of New South Wales |
Eckmann H.,Versicherungsanstalt fur Eisenbahnen und Bergbau |
Hayn D.,Institute for Safety Technology GmbH |
And 3 more authors.
Journal of Telemedicine and Telecare | Year: 2012
Patients with diabetes were enrolled into a telemonitoring programme. They were offered the choice of collecting their health data either by using Near Field Communication (NFC) enabled mobile phones equipped with a dedicated application (App), or by means of a web-browser based user interface (Web). At the end of the study, each patient was categorized as belonging to either the App or Web group, based on the proportion of data they had transmitted using each method. Of the 403 patients, there were 291 in the App group and 112 in the Web group. The two groups were similar in their demographics, except for gender distribution where 68% of men preferred using the App method in contrast to 95% for women (P < 0.001). Kaplan-Meier analysis showed a steady decline of the compliance rate for both groups, at a similar rate during the first year. It also showed a more rapid decline of the compliance rate thereafter for the Web group, which resulted in a significantly higher rate for the App group over the whole observation period (P = 0.03). We conclude that different types of data acquisition technologies may have an important effect on patients' willingness to participate in telehealth programmes in the long-term.
Prokesch R.,Institute for Safety Technology GmbH
IECON Proceedings (Industrial Electronics Conference) | Year: 2013
In this paper manual parallelization is compared with parallelization using the Open Multi-Processing (OpenMP) software framework on an embedded eight core Digital Signal Processor (DSP) developed by Texas Instruments. In order to compare these two methods an in-house developed 2D image registration algorithm is parallelized. Furthermore considerations regarding the parallelization on this DSP are presented. Finally both parallelization methods in respect of performance and real time behavior are measured and analyzed. © 2013 IEEE.
Rass S.,Klagenfurt University |
Konig S.,Institute for Safety Technology GmbH |
Schauer S.,Institute for Safety Technology GmbH
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015
Many decision problems ask for optimal behaviour in (often competitive) situations, where optimality is understood as maximal revenue. The axiomatic approach of von Neumann and Morgenstern establishes the existence of suitable revenue functions, assuming an ordered revenue space. A prominent materialization of this is game theory, where utility functions map actions of several players onto comparable payoffs, typically real numbers. Inspired by an application of that theory to risk management in utility networks, we observed that the usual game-theoretic models are inapplicable due to intrinsic randomness of the effects that an action has. This uncertainty comes from physical and environmental factors that affect the game-play outside of any players influence. To tackle such scenarios, we introduce games in which the payoffs are entire probability distributions (rather than numbers). Towards a sound decision theory, we define a total ordering on a restricted subset of probability distribution functions, and demonstrate how optimal decisions and even basic game theory can be (re)established over abstract revenue spaces of probability distributions. Our results belong to the category of risk control, and are applicable to contemporary security risk management, where decisions must be made under uncertainty and the effects of management actions are almost never deterministic. © Springer International Publishing Switzerland 2015.
Graf R.,Institute for Safety Technology GmbH |
King R.,Institute for Safety Technology GmbH
Archiving 2013 - Final Program and Proceedings | Year: 2013
This paper presents an approach for automatic detection of fingers that mistakenly appear in scans from digitized image collections. Our goal is to create a reliable detection tool that is independent from scan quality, finger sizes, direction, shape, colour and lighting conditions. Modern image processing techniques are applied for edge detection, local image information extraction, and analysis. We employed expert knowledge to determine default parameters of the algorithm, and support customized parameters for specific institutional workflows. Results for three digital collections analysis are presented. Documents with finger artefacts are identified with high reliability and validated by human visual inspection. The proposed method achieves up to 86 percent classification accuracy. © Copyright 2013; Society for Imaging Science and Technology.
Posch C.,Institute for Safety Technology GmbH |
Matolin D.,Institute for Safety Technology GmbH |
Wohlgenannt R.,Institute for Safety Technology GmbH
IEEE Journal of Solid-State Circuits | Year: 2011
The biomimetic CMOS dynamic vision and image sensor described in this paper is based on a QVGA (304× 240) array of fully autonomous pixels containing event-based change detection and pulse-width-modulation (PWM) imaging circuitry. Exposure measurements are initiated and carried out locally by the individual pixel that has detected a change of brightness in its field-of-view. Pixels do not rely on external timing signals and independently and asynchronously request access to an (asynchronous arbitrated) output channel when they have new grayscale values to communicate. Pixels that are not stimulated visually do not produce output. The visual information acquired from the scene, temporal contrast and grayscale data, are communicated in the form of asynchronous address-events (AER), with the grayscale values being encoded in inter-event intervals. The pixel-autonomous and massively parallel operation ideally results in lossless video compression through complete temporal redundancy suppression at the pixel level. Compression factors depend on scene activity and peak at ∼1000 for static scenes. Due to the time-based encoding of the illumination information, very high dynamic range - intra-scene DR of 143 dB static and 125 dB at 30 fps equivalent temporal resolution - is achieved. A novel time-domain correlated double sampling (TCDS) method yields array FPN of < 0.25% rms. SNR is > 56 dB (9.3 bit) for > 10 Lx illuminance. © 2006 IEEE.