2017, Tom 25 Nr 1

Stały URI dla kolekcji

http://hdl.handle.net/11652/3826

Przeglądaj

Przeglądaj 2017, Tom 25 Nr 1 wg Tytuł

Teraz wyświetlane 1 - 4 z 4
  • Pozycja
    Computational Complexity and Numerical Optimization of Adaptive Kuwahara Filter
    (Wydawnictwo Politechniki Łódzkiej, 2017) Bartyzel, Krzysztof
    Adaprive Kuwahara filter produces very interesting results and significantly improves the efficiency and performance of the original algorithm in the context of noise reduction without blurring the edges. This document contains the experimental and theoretical comparison of the computational complexity of the modified algorithm and the consideration of optimization methods.
  • Pozycja
    Digitizing of Analogue Data-Charts from Thermohygrographs
    (Wydawnictwo Politechniki Łódzkiej, 2017) Radon, Jan
    A thermo-hygrograph is a chart recorder that measures and records both temperature and humidity in an analogue form. Converting the results into digital form is necessary to enhance evaluation of measurements archived on paper charts. Coordinate system of the charts is different than common Cartesian rectangular axis. Temperature and relative humidity axis are curved according to recorder pen holder ray. Different chart types, depending on particular thermohygrograph model, were used so far. In every case, however, the axis grid depicted in the background is geometrically precisely defined. So pattern recognition techniques could be used to find its location automatically. Temperature and relative humidity patterns can be retrieved automatically assuming that the color group of plotted lines is different than the background including the grid. Due to storing conditions, manual descriptions, discoloration and other factors, the plotted pattern cannot be unambiguously identified. Many randomly scattered pixels are interpreted as being part of a plotted line. Among additional measures, like excluding particular areas from analysis, manually removing and including points, a vector analysis of retrieved courses makes digitizing more effective.
  • Pozycja
    Generalized Structure of the Algorithm for Automated Detection of Non Relevant and Wrong Information on Web Resources
    (Wydawnictwo Politechniki Łódzkiej, 2017) Dyvak, Mykola; Kovbasistyi, Andrii; Stakhiv, Petro; Lipiński, Piotr
    In this article the algorithm for automated detection of non-relevant or wrong information on websites is introduced. The algorithm extracts the semantic information from the webpage using third party software and compares the semantic information with the reliable resources. Reliable information is identified by the means of majority voting or extracted from reliable databases.
  • Pozycja
    Multidimensional Neo-Fuzzy-Neuron for Solving Medical Diagnostics Tasks in Online-Mode
    (Wydawnictwo Politechniki Łódzkiej, 2017) Mahmoud, Samer Mohamed Kanaan; Perova, Iryna; Pliss, Iryna
    In this paper neuro-fuzzy approach for medical data processing is considered. Special capacities for methods and systems of Computational Intelligence were introduced for Medical Data Mining tasks, like transparency and interpretability of obtained results, ability to classify nonconvex and overlapped classes that correspond to various diagnoses, necessity to process data in online mode and so on. Architecture based on the multidimensional neo-fuzzy-neuron was designed for situation of many diagnoses. For multidimensional neo-fuzzy-neuron adaptive learning algorithms that are a modification of Widrow-Hoff algorithm were introduced. This system was approbate on nervous system diseases data set from University of California Irvine (UCI) Repository and show high level of classification results.