Digitální knihovnaUPCE
 

Fakulta elektrotechniky a informatiky / Faculty of Electrical Engineering and Informatics

Stálý URI pro tuto komunituhttps://hdl.handle.net/10195/3847

Práce obhájené před rokem 2008 jsou uloženy pouze v kolekci Vysokoškolské kvalifikační práce

Procházet

Search Results

Nyní se zobrazuje 1 - 10 z 2393
  • Náhled
    Konferenční objektOmezený přístuppeer-reviewedpostprint
    Design of Data Access Architecture Using ORM Framework
    (IEEE (Institute of Electrical and Electronics Engineers), 2023) Majerík, Filip; Borkovcová, Monika
    Nowadays, various Object-relational mapping frameworks are becoming a key part of computer system architecture. These frameworks provide developers with relatively easy manipulation of data stored in various database systems, even without knowledge of complex database systems. In this article, we have focused on leveraging the benefits of implementing an ORM framework while minimizing the impact on software performance. The design of an intelligent data intermediate layer is described within this paper. This provides optimized communication between the application layer and subsequently the ORM Framework. At the same time, attempts have been made to extend the layer with an additional caching layer, which however proved to be unhelpful for simple SQL queries.
  • Náhled
    Konferenční objektOmezený přístuppeer-reviewedpostprint
    ECG Hearbeat Classification Based on Multi-scale Convolutional Neural Networks
    (Springer Nature Switzerland AG, 2023) Rozinek, Ondřej; Doležel, Petr
    Clinical applications require automating ECG signal processing and classification. This paper investigates the impact of multiscale input filtering techniques and feature map blocks on the performance of CNN models for ECG classification. We conducted an ablation study using the AbnormalHeartbeat dataset, with 606 instances of ECG time series divided into five classes. We compared five multiscale input filtering techniques and four multiscale feature map blocks against a base model and non-multiscale input. Results showed that the combination of mean filter for multiscale input and residual connections for multiscale block achieved the highest accuracy of 64.47%. Residual connections were consistently effective across different filtering techniques, highlighting their potential to enhance CNN model performance for ECG classification. These findings can guide the design of future CNN models for ECG classification tasks, with further experimentation needed for optimal combinations in specific applications.
  • Náhled
    ČlánekOmezený přístuppeer-reviewedpostprint
    Theorems for Boyd-Wong Contraction Mappings on Similarity Spaces
    (MDPI, 2023) Rozinek, Ondřej; Borkovcová, Monika
    In this article, we introduce novel fixed point results for Boyd-Wong-type contraction mappings within the framework of similarity spaces, which have broad practical applications. The development of these results extends the classical theory of Boyd-Wong contractions by exploiting the unique structure and properties of similarity spaces. We provide an in-depth examination of the derived contractions, establishing conditions under which fixed points exist and are unique. In the latter part of the paper, we illustrate the applicability and effectiveness of the proposed results with representative examples.
  • Náhled
    Konferenční objektOmezený přístuppeer-reviewedpostprint
    Automated Dataset Enhancement Using GAN for Assessment of Degree of Degradation around Scribe
    (IEEE (Institute of Electrical and Electronics Engineers), 2023) Doležel, Petr; Pakosta, Marek; Rozsívalová, Veronika; Štursa, Dominik
    Coil coating is a method of applying an organic coating material to a rolled metal strip substrate in a continuous automated process. It is used to provide a high quality, durable finish to a variety of surfaces. The degradation resistance of coil-coated materials is assessed according to European Standard EN 13523-8 by exposing a coil-coated test specimen to a salt fog at a defined temperature for a defined period of time. After this process, a sample is tested according to the International Organisation for Standardisation ISO 4628 standard to determine the degree of degradation. In this study, a GAN-based technique for automated training set enhancement is proposed to assess the degree of degradation around a scribe. The presented technique is capable of enhancing a manually generated dataset of images with synthetic samples to help refine the performance of the area degradation detector.
  • Náhled
    Konferenční objektOmezený přístuppeer-reviewedpostprint
    Object Detection Algorithms - A Review
    (IEEE (Institute of Electrical and Electronics Engineers), 2023) Letavay, Marek; Bažant, Michael; Tuček, Pavel
    Over several decades, traffic engineers have persistently focused on implementing various measures to enhance the safety of drivers and pedestrians amidst the growing traffic on streets worldwide. This pressing concern stems from the continuous growth in vehicles, pedestrians, and road density. However, with the remarkable advancements in computational power and the simultaneous reduction in the cost of technical equipment, the integration of object detection has emerged as a viable solution, even in real-time applications within the realm of traffic engineering. However, it is essential to note that the impact of computer technology extends far beyond mobility-related issues. Its wide-ranging influence has permeated diverse fields of interest, including medicine, where it aids in diagnostics and treatments, face recognition systems used for security, industrial automation processes, and even artistic endeavors, where it has opened new frontiers of creativity. This indicates object detection technology's vast potential and versatility, with its ability to revolutionize numerous sectors and bring about meaningful societal advancements. This review outlines the chronological progression of the essential YOLO algorithm, highlighting its significant advancements and developments over the years.
  • Náhled
    ČlánekOmezený přístuppeer-reviewedpostprint
    Verification and estimation of uncertainties of Tobias Mayer's 18th century astronomical observations
    (Elsevier Science BV, 2023) Marek, Jaroslav; Tuček, Jiří
    Perceiving the uncertainty of the measurement has been changing over the past centuries, reflecting the advancement in the experimental techniques, the urge for reliable and reproducible measurement methodology, and development of mathematical data processing and evaluating algorithms. From the historical perspective, the concepts of considering the measurement uncertainty were firstly introduced with geographic and cartographic measurements. In this context, the works of Tobias Mayer on lunar landscape measurements are widely highlighted which, at that time, presented innovative approaches in data processing with the method of averages and pioneeringly addressed the issue of measurement error. In this study, we analyze in details the Mayer's set of 27 non-linear equations with 3 unknown parameters and discuss the effect of Mayer's linearization and subsequent mathematical procedures on the accuracy of the parameter values in contrast with the results from rigorous treatment of non-linear regression model involving the least-square method. In particular, we compare the values of the unknown parameters and their uncertainties in several variants in the linearized and nonlinearized model, providing monitoring of a small deviation of the Mayer's linearization. The results, presented here, show that despite the conceptual and computational simplification of the Mayer's method, such an approach to data processing can be exploited, with an acceptable level of accuracy, in several practical situations even today.
  • Náhled
    Konferenční objektOmezený přístuppeer-reviewedpostprint
    The Exact Solution of Vehicle Routing Problem by Mixed Integer Linear Programming in Matlab
    (Czech Society for Operations Research, 2023) Zahrádka, Jaromír; null, null
    This contribution comes up with a specific solution of the vehicle routing problem. The driver has to deliver the goods from the central warehouse to n customers as efficiently as possible. Each customer has ordered goods that fill a certain number of containers. Each customer point of delivery is given by GPS coordinates. The objective of the solution is to select the number of vehicles and their routes between customers in such a way that the total travel time, including the time for unloading the goods, is as short as possible. Each delivery point is visited only once by one of the vehicles. All used vehicles have a pre-limited capacity of containers. All vehicles return to the central warehouse. In this contribution, the algorithm of the exact solution of the vehicle routing problem was created, which can be used in general for any number n of customers. The algorithm is implemented in Matlab code.
  • Náhled
    ČlánekOmezený přístuppeer-reviewedpostprint
    Survey of Point Cloud Registration Methods and New Statistical Approach
    (MDPI, 2023) Marek, Jaroslav; Chmelař, Pavel
    The use of a 3D range scanning device for autonomous object description or unknown environment mapping leads to the necessity of improving computer methods based on identical point pairs from different point clouds (so-called registration problem). The registration problem and three-dimensional transformation of coordinates still require further research. The paper attempts to guide the reader through the vast field of existing registration methods so that he can choose the appropriate approach for his particular problem. Furthermore, the article contains a regression method that enables the estimation of the covariance matrix of the transformation parameters and the calculation of the uncertainty of the estimated points. This makes it possible to extend existing registration methods with uncertainty estimates and to improve knowledge about the performed registration. The paper's primary purpose is to present a survey of known methods and basic estimation theory concepts for the point cloud registration problem. The focus will be on the guiding principles of the estimation theory: ICP algorithm; Normal Distribution Transform; Feature-based registration; Iterative dual correspondences; Probabilistic iterative correspondence method; Point-based registration; Quadratic patches; Likelihood-field matching; Conditional random fields; Branch-and-bound registration; PointReg. The secondary purpose of this article is to show an innovative statistical model for this transformation problem. The new theory needs known covariance matrices of identical point coordinates. An unknown rotation matrix and shift vector have been estimated using a nonlinear regression model with nonlinear constraints. The paper ends with a relevant numerical example.
  • Náhled
    Konferenční objektOmezený přístuppeer-reviewedpostprint
    Passage Detection of a Train via a Reference Point
    (Springer, 2023) Rejfek, Luboš; Pidanič, Jan; Štursa, Dominik; Nguyen, Tan N.; Tran, Phuong T.; Němec, Zdeněk; Zálabský, Tomáš
    A reference point detection system for position validation of a mobile object was developed for verification of experiments. The detection is based on a classic image processing algorithm and a processing algorithm using neural networks. Both approaches are compared. High-precision concept of the system is based on a camera sensor and automatic processing of video frames for position evalua-tion. The designed system was tested on a real application proving correct operation.
  • Konferenční objektOmezený přístuppeer-reviewedpostprint
    Cybersecurity of Sensors on Smart Vehicles: Review of Threats and Solutions
    (IEEE (Institute of Electrical and Electronics Engineers), 2023) Putro, Prasetyo Adi Wibowo; Amelia, Fetty; Pidanič, Jan; Suhartanto, Heru; Rahardjo, Imam Arif; Imandeka, Ejo
    The use of sensors in smart vehicles brings benefits and vulnerabilities. Different kinds of sensors in smart vehicles are vulnerable to cyber-attack. Until now, the investigation of challenges and solutions for in-vehicle cybersecurity hasn’t discussed various sensor objects and their correlation. In this study, we studied the cyber security problems of sensors in smart vehicles and how to overcome them. The research was designed as Systematic Literature Review (SLR) using the Kitchenham methodology with modification in the filtering phase using the artificial intelligence application, Elicit, to identify the problems, conclusions, and methodology description. Seventeen publications from 2016 until 2023 were gained from five databases. As a result, we find that the most discussed object related to cybersecurity sensors on smart vehicles are Electronic Control Units. Spoofing and jamming is still the most addressed threat, and machine learning is the most utilized solution to be implemented in detection systems. Advanced detection systems are incorporating updated attack models. We also suggest using updated attack models and machine learning algorithms to ensure the safety and security of smart vehicle technology. All identified sensor technology correlated using mind maps under the Intelligent Transport System theory.