This thesis focuses on the development of a system for recognizing graphic images using artificial neural networks, particularly the Hopfield and ART-1 networks. The aim was to design and implement these networks to improve pattern recognition accuracy. Both architectures were implemented and thoroughly tested, with analyses conducted on their ability to restore graphic patterns and their performance depending on the number of neurons and network capacity. The Hopfield network demonstrated a fast convergence time but limited accuracy with more complex datasets, while the ART-1 network, due to its high capacity, showed potential in creating numerous categories. However, the accuracy of ART-1 was low, indicating the need for further optimization. This study highlights the need for continued research and improvements in parameters such as vigilance to enhance the efficiency and applicability of these networks in graphic analysis tasks.

Virtual reality (VR) is an extremely rapidly developing field of technology that is having a significant impact on our perception and the way we perceive the world. However, the identification of hidden objects for visual perception in VR remains an issue that requires deeper understanding and research. The present work aims to systematize knowledge about the processes of visual perception in virtual reality and the identification of objects that remain hidden or are difficult to detect in this environment. The work is based on experimental studies conducted to identify factors that affect our ability to detect objects in a virtual environment and to evaluate the impact of different visualization techniques on the perception of hidden objects. The results of these studies show the important role of visual stimuli when using VR. Despite the subjects’ indication that the research task had a balanced degree of difficulty, and that the auditory stimulus allowed for the accurate location of obstacles, participants found navigating the virtual space frustrating and aggravating.

The purpose of the study was to systematize knowledge about the effect of the country of origin of the product and to determine the impact of the importance of the country of origin of the product on the purchasing decisions of a selected group of foreigners living in Poland. Chapter one of the work characterizes the purchasing decisions of consumers and the determinants of these decisions. Chapter two of the work describes the concepts of the effect of the country of origin of the product as a factor affecting purchasing decisions. The third chapter of the work presents the results and analyzes the research conducted on how information about the country of origin of the product affects the purchasing decisions of foreigners living in Poland. The methods used in the work are the descriptive method, as well as the survey method and the method of own research. The information contained in the various chapters of the work was based on Polish and English-language literature on the subject, as well as on scientific articles and materials available on the Internet.

The aim of this work is to explore the possibilities of minimizing the operational vertices of algorithmic state machines. The author presents the existing types of these machines and various methods of their optimization - incl. the classical method, as well as the benefits of conducting the optimization. A solution is proposed based on a combination of a few simple methods of optimizing conditional vertices. On its basis, the software implementing this theory is developed with the help of the C# language. This software is adapted to operate on ASMCreator files, which is a popular program for creating flowcharts and compiling them into text form. The operation of the program has been tested on numerous author’s cases and seems to be stable and reliable, but it is not certain that the program generates the optimal result for each given flowchart. In some examples, it is possible to see unrealized minimization due to the lack of implementation of the conditional vertex minimization algorithm, the idea of which has not been fully explored in this paper.