Fraunhofer-Institut für Produktionsanlagen und Konstruktionstechnik IPK

Filters

58
1
Reset filters

Settings
Now showing 1 - 10 of 58
  • Publication
    Evolutionary multi-objective optimization of particle swarm optimizers
    ( 2007)
    Veenhuis, C.
    ;
    Köppen, M.
    ;
    Vicente-Garcia, R.
    One issue in applying Particle Swarm Optimization (PSO) is to And a good working set of parameters. The standard settings often work sufficiently but don't exhaust the possibilities of PSO. Furthermore, a trade-off between accuracy and computation time is of interest for complex evaluation functions. This paper presents results for using an EMO approach to optimize PSO parameters as well as to And a set of trade-offs between mean fitness and swarm size. It is applied to four typical benchmark functions known from literature. The results indicate that using an EMO approach simplifies the decision process of choosing a parameter set for a given problem.
  • Publication
    A framework for the adaptation of image operators
    ( 2007)
    Köppen, M.
    ;
    Vicente-Garcia, R.
    A framework was presented, which allows for the design of texture filters for fault detection (two class problem). The framework is based on the 2D-Lookup algorithm, where two filter output images are used as input. "figure presented" "figure presented" The approach can be applied to a large class of texture analysis problems. The results, obtained without "human intervention," are ready-to-use texture filters. Also, they can be tuned in order to obtain even more better results, or combined in a superposed inspection system. The following are our experiences during the use of the system. (i) The framework was able to design texture filters with good or very good performance. (ii) The goal image matched the fault region quite satisfactorily. (iii) Bordering regions should be neglected for fitness evaluation. (iv) The framework was able to design filters for the detection of noncompact fault regions and fault regions with varying appearance. (v) The designed filters may b e subjected to further improvements by the user. "figure presented" Improvements of the whole architecture were considered as well: one is based on an evaluation of the 2D-Lookup matrix by neural networks in order to get a more comprehensive solution for a given texture filtering problem, the other for extending the application scope to low-contrast texture fault processing, that is, faults which are hard to separate from the background texture. The second extension of the framework is a two-stage one, based on 2D histogram lookup and consecuting 2D-Lookup adaptation.
  • Publication
    Data swarm clustering
    ( 2006)
    Veenhuis, C.
    ;
    Köppen, M.
  • Publication
    Tiny GAs for image processing applications
    ( 2006)
    Köppen, M.
    ;
    Franke, K.
    ;
    Vicente-Garcia, R.
    The expedience of today's image-processing applications is no longer based on the performance of a single algorithm alone. These systems appear to be complex frameworks with a lot of sub-tasks that are solved by specific algorithms, adaptation procedures, data handling, scheduling, and parameter choices. The venture of using computational intelligence (CI) in such a context, thus, is not a matter of a single approach. Among the great choice of techniques to inject CI in an image-processing framework, the primary focus of this presentation will be on the usage of so-called Tiny-GAs. This stands for an evolutionary procedure with low efforts, i.e. small population size (like 10 individuals), little number of generations, and a simple fitness. Obviously, this is not suitable for solving highly complete: optimization tasks, but the primary interest here is not the best individual's fitness, but the fortune of the algorithm and its population, which has just escaped the Monte-Carlo domain after random initialization. That this approach can work in practice will be demonstrated by means of selected image-processing applications, especially in the context of linear regression and line fitting; evolutionary post processing of various clustering results, in order to select a most suitable one by similarity; and classification by the fitness values obtained after a few generations.
  • Publication
    Entwicklung eines lernfähigen Bildverarbeitungssystems unter Einsatz von Verfahren des Soft Computing
    (Fraunhofer IRB Verlag, 2006)
    Köppen, M.
    ;
    Seliger, G.
    Moderne Bildverarbeitungssysteme weisen als Folge wachsender Anforderungen aus ihren Einsatzgebieten eine immer höher werdende Komplexität auf. Manuelle Anpassungen für eine optimale Systemleistung sind schon lange nicht mehr ausreichend. Moderne lernfähige Verfahren des Soft Computing ermöglichen bereits im Prozess der Gestaltung und Entwicklung einer Bildverarbeitungsanwendung eine optimale Anpassung für den späteren Einsatz. Der in dieser Arbeit erarbeitete Lösungsansatz besteht in der Entwicklung eines flexiblen, lernfähigen Systems zur effizienten Konfiguration von Bildverarbeitungssystemen als entscheidende Tätigkeit des Entwicklers und unter Beachtung der Optimierungsziele hohe Systemleistung, hoher Systemdurchsatz, hohe Flexibilität der Gestaltung und hoher Grad an Autonomie. In der Aufarbeitung der theoretischen Grundlagen des Soft Computing werden die Möglichkeiten, aber auch die Grenzen des Einsatzes dieser Methoden aufgezeigt. Die Implikationen für die Entwicklung von Bildverarbeitungssystemen in diesen theoretischen Grenzen werden in Form von Design-Vorschriften dargestellt. Am Beispiel eines Systems zur optimalen Anpassung von Texturfiltern wird exemplarisch eine darauf aufbauende Systemlösung konzipiert und umgesetzt. Die Ergebnisse aus dem Einsatz dieses Systems bestätigen die praktische Umsetzbarkeit des gewählten Lösungsansatzes.
  • Publication
    General chairs' welcome message
    ( 2005)
    Köppen, M.
    ;
    Corne, D.W.
    ;
    Abraham, A.
  • Publication
    Improvement of a face detection system by evolutionary multi-objective optimization
    ( 2005)
    Verschae, R.
    ;
    Solar, J.R. del
    ;
    Köppen, M.
    ;
    Garcia, R.V.
    This paper presents the application of evolutionary multi-objective optimization (EMO) to the improvement of a face detection system. The face detection system is based on the boosted cascade system, and analyzes image positions on different scales in a three-step-procedure. Based on threshold settings, the algorithm decides whether to continue with the test on a finer scale at the current position. Thus, the thresholds for all scales and stages have a major influence on the performance of the system, and become the subject of the evolutionary optimization according to three objectives: low false positive rate, high detection rate and low processing time. The used EMO is the extension of the Standard Genetic Algorithm to the EMO case by using Fuzzy Pareto Dominance as a meta-heuristic. The application of this EMO to the face detection system is explored and discussed using images from a standard face detection benchmark dataset. From the runtime analysis of the EMO it c an be concluded that the algorithm reliably approaches the Pareto set of the problem.
  • Publication