Перегляд за автором "Rybalka, A. I."

Зараз показано 1 - 4 з 4
  • Зображення мініатюри
    Публікація
    Assessment of the effectiveness of laser-acoustic transformation
    (Technical University of Sofia, 2019) Rybalka, A. I.; Megel, Y. E.; Kalimanova, I.
    For some biotechnological technologies, there is a problem of separating living cells or destroying their membranes in order to obtain individual elements of the inner structure of the cell. This is usually done using microsurgical operations. In practice, the method of obtaining individual cell fragments without surgery using an acoustic wave is of interest. In this paper, a mathematical model and the results of modeling the process of the formation of an acoustic pressure wave in a liquid biological solution for the destruction of the membrane of a biological object is considered.
  • Немає доступних мініатюр
    Публікація
    Cattle breed identification and live weight evaluation the basis of machine learning and computer vision
    (Національний університет "Запорізька політехніка", 2020) Megel, Y. E.; Rudenko, O. G.; Bezsonov, O. O.; Rybalka, A. I.
    The problem of the cow’s live weight estimation is considered. A convolutional neural network based method for animal recognition and its breed identification in combination with epipolar geometry approach for object’s size measurement is proposed. Information regarding animal’s size and its breed is further used for LW estimation by multilayer perceptron based predictive model. This approach can be used to replace traditional methods of direct observation and measurement. The proposed system can be widely used in the management of a modern farm. Accuracy and performance of the proposed method has been tested with the participation of the experts.
  • Зображення мініатюри
    Публікація
    Measuring multimedia content proximity via artificial intelligence methods
    (Technical University of Sofia, 2020) Rybalka, A. I.; Megel, Y. E.; Mikhnova, O.
    The paper describes existing and novel approaches to word sequence similarity measuring which uses image processing procedures to reveal proximity relations between words and their meaning. This is implemented due to the fact the each word sequence can be associated with a particular pictorial domain, and thus any textual information can be potentially presented via an image series. Theoretical results of rule based approach, genetic algorithms and neural networks application are analyzed in terms of solving the problem under study. The proposed information processing technology may be useful for automatic translation applications, known as CAT-tools by now, search engine optimization, etc.
  • Зображення мініатюри
    Публікація
    Modelling of an automated food quality assessment system based on fuzzy inference
    (Харківський національний університет радіоелектроніки, 2020) Rybalka, A. I.; Kutsenko, A. S.; Kovalenko, S. V.
    The purpose of this study is to create a methodology for developing an automated system for assessing the quality of food products based on a comprehensive quality indicator and the use of fuzzy logic theory, namely, fuzzy inference. In our opinion, such an approach to quality assessment can reduce the subjective component that has a significant impact on making a final decision. The system, built on a given algorithm, allows us to assess the quality of food products, taking into account the data of laboratory studies on measurable quality indicators and expert opinions on difficult to measure indicators.