004 Datenverarbeitung; Informatik
Filtern
Erscheinungsjahr
- 2009 (13) (entfernen)
Dokumenttyp
- Ausgabe (Heft) zu einer Zeitschrift (6)
- Dissertation (3)
- Diplomarbeit (2)
- Bachelorarbeit (1)
- Masterarbeit (1)
Sprache
- Englisch (13) (entfernen)
Schlagworte
- Bildverarbeitung (2)
- Akzeptanz (1)
- Automatische Klassifikation (1)
- BPMS (1)
- Bildsegmentierung (1)
- Bluetooth (1)
- CSCW (1)
- Coloskopie (1)
- Computersimulation (1)
- Darmpolyp (1)
The STOR project aims at the development of a scientific component system employing models and knowledge for object recognition in images. This interim report elaborates on the requirements for such a component system, structures the application area by identifying a large set of basic operations, and shows how a set of appropriate data structures and components can be derived. A small case studies exemplifies the approach.
Hybrid systems are the result of merging the two most commonly used models for dynamical systems, namely continuous dynamical systems defined by differential equations and discrete-event systems defined by automata. One can view hybrid systems as constrained systems, where the constraints describe the possible process flows, invariants within states, and transitions on the one hand, and to characterize certain parts of the state space (e.g. the set of initial states, or the set of unsafe states) on the other hand. Therefore, it is advantageous to use constraint logic programming (CLP) as an approach to model hybrid systems. In this paper, we provide CLP implementations, that model hybrid systems comprising several concurrent hybrid automata, whose size is only straight proportional to the size of the given system description. Furthermore, we allow different levels of abstraction by making use of hierarchies as in UML statecharts. In consequence, the CLP model can be used for analyzing and testing the absence or existence of (un)wanted behaviors in hybrid systems. Thus in summary, we get a procedure for the formal verification of hybrid systems by model checking, employing logic programming with constraints.
The lack of a formal event model hinders interoperability in distributed event-based systems. Consequently, we present in this paper a formal model of events, called F. The model bases on an upper-level ontology and pro-vides comprehensive support for all aspects of events such as time and space, objects and persons involved, as well as the structural aspects, namely mereological, causal, and correlational relationships. The event model provides a flexible means for event composition, modeling of event causality and correlation, and allows for representing different interpretations of the same event. The foundational event model F is developed in a pattern-oriented approach, modularized in different ontologies, and can be easily extended by domain specifific ontologies.
Avoidance of routing loops
(2009)
We introduce a new routing algorithm which can detect routing loops by evaluating routing updates more thoroughly. Our new algorithm is called Routing with Metric based Topology Investigation (RMTI), which is based on the simple Routing Information Protocol (RIP) and is compatible to all RIP versions. In case of a link failure, a network can reorganize itself if there are redundant links available. Redundant links are only available in a network system like the internet if the topology contains loops. Therefore, it is necessary to recognize and to prevent routing loops. A routing loop can be seen as a circular trace of a routing update information which returns to the same router, either directly from the neighbor router or via a loop topology. Routing loops could consume a large amount of network bandwidth and could impact the endtoend performance of the network. Our RMTI approach is capable to improve the efficiency of Distance Vector Routing.
Software is vital for modern society. The efficient development of correct and reliable software is of ever-growing importance. An important technique to achieve this goal is deductive program verification: the construction of logical proofs that programs are correct. In this thesis, we address three important challenges for deductive verification on its way to a wider deployment in the industry: 1. verification of thread-based concurrent programs 2. correctness management of verification systems 3. change management in the verification process. These are consistently brought up by practitioners when applying otherwise mature verification systems. The three challenges correspond to the three parts of this thesis (not counting the introductory first part, providing technical background on the KeY verification approach). In the first part, we define a novel program logic for specifying correctness properties of object-oriented programs with unbounded thread-based concurrency. We also present a calculus for the above logic, which allows verifying actual Java programs. The calculus is based on symbolic execution resulting in its good understandability for the user. We describe the implementation of the calculus in the KeY verification system and present a case study. In the second part, we provide a first systematic survey and appraisal of factors involved in reliability of formal reasoning. We elucidate the potential and limitations of self-application of formal methods in this area and give recommendations based on our experience in design and operation of verification systems. In the third part, we show how the technique of similarity-based proof reuse can be applied to the problems of industrial verification life cycle. We address issues (e.g., coping with changes in the proof system) that are important in verification practice, but have been neglected by research so far.
Entwicklung eines Regelungsverfahrens zur Pfadverfolgung für ein Modellfahrzeug mit Sattelanhänger
(2009)
Neben der fortschreitenden Automatisierung im innerbetrieblichen Warenverkehr ist auch die Automatisierung in ausgewählten Bereichen des ausserbetrieblichen Waren- und Güterverkehrs erstrebenswert. Durch den Einsatz von fahrerlosen Lkw-Gespannen auf Speditionshöfen kann die ökonomische Effizienz, der dort anfallenden Abläufe, erheblich erhört werden. Insbesondere werden dazu präzise Regelungsverfahren benötig, die auch für Sattelzüge ein exaktes Abfahren vorgegebener Wege gewährleisten. Das allgemeine Ziel dieser Arbeit ist die Adaption und Evaluation eines Regelverfahrens zur Pfadverfolgung für Sattelzuggespanne. Die Unterschiede im kinematischen Verhalten zwischen LKW mit einem einachsigen Starrdeichselanhänger und Sattelzügen herausgearbeitet werden. Im Weiteren werden die charakteristischen kinematischen Eigenschaften von Sattelzügen bei der Adaption eines Regelverfahrens berücksichtigt, das zunächst speziell für Fahrzeuge mit Starrdeichselanhänger konzipiert wurde. Das Regelungsverfahren zur Pfadverfolgung muss sowohl für vorwärts als auch rückwärtsgerichtete Fahrmanöver geeignet sein. Das Regelungsverfahren wird als abgeschlossene Komponente in die Steuersoftware eines Modellfahrzeugs integriert. Dazu werde für die Geometrie des Modellfahrzeugs spezifische mit dem Ziel, Grenzen möglicher Regelabweichungen zu bestimmen. Die Arbeit dokumentiert darüber hinaus die zentralen Softwarekomponenten des implementierten Regelungsverfahrens
Die Koloskopie ist der Goldstandard zur Aufspürung von gefährlichen Darmpolypen, die sich zu Krebs entwickeln können. In einer solchen Untersuchung sucht der Arzt in den vom Endoskop gelieferten Bildern nach Polypen und kann diese gegebenenfalls entfernen. Um den Arzt bei der Suche zu unterstützen, erforscht die Universität Koblenz-Landau zur Zeit Methoden, die zur automatischen Detektion von Polypen auf endoskopischen Bildern verwendet werden können. Wie auch bei anderen Systemen zur Mustererkennung werden hierzu zunächst Merkmale aus den Bildern extrahiert und mit diesen ein Klassifikator trainiert. Dieser kann dann für die Klassifikation von ihm unbekannten Bildern eingesetzt werden. In dieser Arbeit wurde das vorhandene System zur Polypendetektion um Merkmalsdetektoren erweitert und mit den bereits vorhandenen verglichen. Implementiert wurden Merkmale basierend auf der Diskreten Wavelet-Transformation, auf Grauwertübergangsmatrizen und auf Local Binary Patterns. Verschiedene Modifikationen dieser Merkmale wurden getestet und evaluiert.
This Thesis contributes by reporting on the current state of diffusion of collaboration information technology (CIT). The investigation concludes, with a high degree of certainty, that today we have a "satisfactory" diffusion level of some level-A CITs (mostly e-Mail, distantly followed by Audio Conferencing), and a "dissatisfactory" diffusion level of higher-level CITs (i.e. those requiring significant collaboration and cooperation among users, like Meeting Support Systems, Group Decision Support Systems, etc.). The potential benefits of the latter seem to be far from fully realised due to lack of user acceptance. This conclusion has gradually developed along the research cycle " it was suggested by Empirical Study I, and tested through Empirical Studies II and III. An additional, unplanned and rather interesting, finding from this study has been the recognition of large [mostly business] reporting on numerous Web 2.0 user-community produced collaboration technologies (most of them belonging to the category of "social software") and their metamorphosis from autonomous, "bottom-up" solutions into enterprise-supported infrastructures. Another contribution of this Thesis " again suggested by Empirical Study I, and tested through Empirical Studies II and III " pertains to the "process structure" of CIT diffusion. I have found that collaboration technology has historically diffused following two distinct (interdependent but orthogonal) diffusion paths " top-down (authority-based) and bottom-up. The authority-based diffusion path seems to be characterised by efforts aimed at "imposing" technologies on employees, the primary concern being to make sure that technology seamlessly and easily integrates into the organisational IT infrastructure. On the other hand, the bottom-up diffusion trail seems to be successful. The contribution of this investigation may be summarised as threefold: 1. This investigation consolidates most of the findings to date, pertaining to CIT adoption and diffusion, which have been produced by the CIT research community. Thus, it tells a coherent story of the dynamics of the community focus and the collective wisdom gathered over a period of (at least) one decade. 2. This work offers a meaningful framework within which to analyse existing knowledge " and indeed extends that knowledge base by identifying persistent problems of collaboration technology acceptance, adoption and diffusion. These problems have been repeatedly observed in practice, though the pattern does not seem to have been recognised and internalised by the community. Many of these problems have been observed in cases of CIT use one decade ago, five years ago, three years ago, and continue to be observed today in structurally the same form despite what is unarguably "rapid technological development". This gives me reason to believe that, at least some of the persistent problems of CIT diffusion can be hypothesised as "determining factors". My contribution here is to identify these factors, discuss them in detail, and thus tackle the theme of CIT diffusion through a structured historical narrative. 3. Through my contribution (2) above, I characterise a "knowledge-action gap" in the field of CIT and illuminate a potential path through which the research community might hope to bridge this gap. The gap may be operationalised as cognitive distance between CIT "knowledge" and CIT "action".
In dieser Arbeit werden drei Verfahren zur Objektentfernung aus Bildern einander gegenübergestellt. Zwei der ausgewählten Verfahren stammen aus dem Bereich der sogenannten Inpainting-Verfahren, während das dritte dem Forschungsgebiet der medizinischen Bildverarbeitung entnommen ist. Die Evaluation dieser Verfahren zeigt ihre jeweiligen Vor- und Nachteile auf und prüft ihre Anwendbarkeit auf das spezifische Problem, ein Farbkalibriermuster aus strukturdominierten Bildern zu entfernen. Auf der Grundlage dieser Eigenschaften werden abschließend mehrere Erweiterungen vorgestellt, die eine verbesserte Anwendbarkeit auf das gestellte Problem erreichen.
In recent years, traceability has been more and more universally accepted as being a key factor for the success of software development projects. However, the multitude of different, not well-integrated taxonomies, approaches and technologies impedes the application of traceability techniques in practice. This paper presents a comprehensive view on traceability, pertaining to the whole software development process. Based on graph technology, it derives a seamless approach which combines all activities related to traceability information, namely definition, recording, identification, maintenance, retrieval, and utilization in one single conceptual framework. The presented approach is validated in the context of the ReDSeeDS-project aiming at requirements-based software reuse.