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
47 výsledky
Search Results
Konferenční objektpeer-reviewedpostprint Omezený přístup ECG Hearbeat Classification Based on Multi-scale Convolutional Neural Networks(Springer Nature Switzerland AG, 2023) Rozinek, Ondřej; Doležel, PetrClinical 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.Konferenční objektpeer-reviewedpostprint Omezený přístup 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, DominikCoil 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.Konferenční objektpeer-reviewedpostprint Omezený přístup Comparison of Geospatial Trajectory Clustering and Feature Trajectory Clustering for Public Transportation Trip Data(Springer Nature Switzerland AG, 2023) Cogollos Adrian, Hector; Baruque Zanon, Bruno; Porras Alfonso, Santiago; Doležel, PetrOne of the techniques for the analysis of travel patterns on a public transport network is the clustering of the users movements, in order to identify movement patterns. This paper analyses and compares two different methodologies for public transport trajectory clustering: feature clustering and geospatial trajectory clustering. The results of clustering trip features, such as origin, destination, or distance, are compared against the clustering of travelled trajectories by their geospatial characteristics. Algorithms based on density and hierarchical clustering are compared for both methodologies. In geospatial clustering, different metrics to measure distances between trajectories are included in the comparison. Results are evaluated by analysing their quality through the silhouette coefficient and graphical representations of the clusters on the map. The results show that geospatial trajectory clustering offers better quality than feature trajectory clustering. Also, in the case of long and complete trajectories, density clustering using edit distance with real penalty distance outperforms other combinations.Konferenční objektpeer-reviewedpostprint Omezený přístup Object detection for robotic grasping using a cascade of convolutional networks(IEEE (Institute of Electrical and Electronics Engineers), 2023) Rais, Vítek; Doležel, PetrRobot guidance in industry is a significant issue that needs to be dealt with in modern manufacturing facilities. One of the common tasks in this area is the pick and place problem. For proper implementation of an automatic pick and place application using a robotic arm for object grasping, it is necessary to detect the accurate pose of the objects of interest. In this contribution, a novel engineering approach to object positioning, based on image processing is proposed. In this approach, the operation is composed of a cascade of convolutional neural networks. This cascade consists of 2 different types of networks. The first one is the object detection network called YOLOv5. It is used to process the raw image data from the scene to provide precise localization and determine the position of the objects of interest. After that, crops of the detected objects are created and processed by the second neural network, namely EfficientNet. This classification network is used to determine the rotation angle of the detected objects. The proposed approach provides a precision rate of 0.997 and a recall rate of 0.999 for locating and determining the correct position. For angle classification, EfficientNet provides an accuracy of 0.951. All tests are performed on the testing set of the legitimate positioning problem.Konferenční objektpeer-reviewedpostprint Omezený přístup Multi-Scale Neural Model for Tool-Narayanaswamy-Moynihan Model Parameter Extraction(Springer Nature Switzerland AG, 2023) Pakosta, Marek; Doležel, Petr; Svoboda, Roman; Baruque Zanon, BrunoGlass transitions are an important phenomenon in amor phous materials with potential for various applications. The Tool-Narayanaswamy-Moynihan (TNM) model is a widely used empirical model that describes the enthalpy relaxation behavior of these materials. However, determining the appropriate values for its parameters can be challeng ing. To address this issue, a multi-scale convolutional neural model is pro posed that can accurately predict the TNM parameters directly from the set of differential scanning calorimetry curves, experimentally measured using the sample of the considered amorphous material. The resulting Mean Absolute Error of the model over the test set is found to be 0.0252, indicating a high level of accuracy. Overall, the proposed neural model has the potential to become a valuable tool for practical application of the TNM model in the glass industry and related fields.Konferenční objektpeer-reviewedpostprint Omezený přístup Employing Quantile and Probability Plots for Comparing and Assessing Goodness of Fit for Stochastic Models of the DCT Coefficients of Lossy Compressed Images(IEEE (Institute of Electrical and Electronics Engineers), 2023) Kotov, Dmytro; Fedorov, Oleksii; Omelchenko, Anatolii; Pidanič, Jan; Doležel, PetrThe paper employs probability plots to performgoodness of fit tests for AC DCT coefficients of compressed images. Attention is paid to various probabilistic models of the DCT coefficients, conventional and rarely used. The Laplacian, generalized Gaussian and doubly Gamma distributions comprise the list. A variation of the method of moments, which involves the 2nd and 4th sample moments as well as Sheppard’s corrections to estimate the shape and scale parameters of the distributions is used. Two types of images are considered, namely, texturelike images and those possessing vast regions of monotonicity. Special effort has been put into adjusting the apparatus of the probability plots to make them suitable for dealing with discrete data, in our case with the quantized DCT coefficients of lossy compressed images. As the source of the DCT coefficients, JPEG images are used. This does not lead us to a significant loss of generality: conclusions drawn in this paper remain applicable to a broad variety of formats of lossy compressed images.Konferenční objektpeer-reviewedpostprint (accepted version) Otevřený přístup Design and Implementation of Probabilistic Methods for Spectrum Sensing in Cognitive Radios(IEEE (Institute of Electrical and Electronics Engineers), 2022) Ponomarov, Andrii; Ivanenko, Stanislav; Fedorov, Oleksii; Bezruk, Valeriy; Pidanič, Jan; Doležel, PetrThe paper deals with new unconventional methods of detecting unoccupied frequency channels in cognitive radios. The main feature of these methods consists in their ability of detecting unknown signals in the presence of noise under the condition of a priori uncertainty. It makes it possible to increase the efficiency of detecting unoccupied frequency channels in cognitive radios due to the fact that these methods track changes in the probabilistic properties of observations. During the course of spectrum sensing of the frequency range, the detected signals are divided into known (classified training samples of which are available in the system) and unknown ones. Application of methods for recognizing specified signals in the presence of unknown signals makes it possible to simultaneously avoid the erroneous occupation of a frequency channel by a secondary user, in the case when previously unregistered signal occurs, and also refresh the cognitive radio database. To detect unknown signals, only information about probabilistic characteristics of the channel noise is used.Konferenční objektpeer-reviewedpostprint (accepted version) Otevřený přístup Grasping Point Detection Using Monocular Camera Image Processing and Knowledge of Center of Gravity(Springer Nature Switzerland AG, 2022) Štursa, Dominik; Doležel, Petr; Honc, DanielThe ability to grasp objects is one of the basic functions of modern industrial robots. In this article, the focus is placed on a system for processing the image provided by a robot visual perception system leading to the detection of objects grasping points. The proposed processing system is based on a multi-step method using convolutional neural networks (CNN). The first step is to use the first CNN to transform the input image into a schematic image with labeled objects centers of gravity, which then serves as a supporting input to the second CNN. In this second CNN, original input and supporting input images are used to obtain a schematic image containing the grasping points of the objects. This solution is further compared with a network providing grasping points directly from the input image. As a result, the proposed method provided a 0.7% improvement in the average intersection over union for all of the models.Konferenční objektpeer-reviewedpostprint (accepted version) Omezený přístup Classification of Polymers Based on the Degree of Their Transparency in SWIR Spectrum(Springer Nature Switzerland AG, 2022) Štursa, Dominik; Kopecký, Dušan; Roleček, Jiří; Doležel, Petr; Baruque Zanon, BrunoDetection, classification and sorting of polymeric particles is a common task required in recycling industry. In the proposed work, an innovative method for detection of polymeric particles and their classification is introduced. The method is based on evaluation of images of polymeric particles, obtained from short-wavelength infrared (SWIR) camera, by convolutional neural network (CNN). Compared to conventionally used spectroscopes or hyper-spectral imaging, this method utilizes single wavelength (1 050 nm) and a degree of polymer transparency serves as the main descriptor. Five different polymers (ABS, ABS-T, Nylon, PETG, PLA) in form of regular blocks (size 15 × 15 × 0.3 mm) were used in the experiment. In total 203 images (size 288 × 288 px) were prepared for CNN training and 67 for testing. Scalable ASP U-Net was tested in 6 combinations and their outputs were compared. According to used intersection over union metrics over all outputs, the topology with 64 filters and depth of 3 exhibited the best results.Konferenční objektpeer-reviewedpostprint (accepted version) Otevřený přístup Suitable ASP U-Net training algorithms for grasping point detection of nontrivial objects(IEEE (Institute of Electrical and Electronics Engineers), 2022) Doležel, Petr; Štursa, Dominik; Kopecký, DušanRobotic manipulation with nontrivial or irregular objects, which provide various types of grasping points, is of both academic and industrial interest. Recently, a powerful data-driven ASP U-Net deep neural network has been proposed to detect feasible grasping points of manipulated objects using RGB data. The ASP U-Net showed the ability to detect feasible grasping points with exceptional accuracy and more than acceptable inference times. So far, the network has been trained using an Adam optimizer only. However, in order to optimally utilize the potential of ASP U-Net, it was necessary to perform a systematic investigation of suitable training algorithms. Therefore, the aim of this contribution was to extend the impact of ASP U-Net by recommending suitable training algorithms and their parameters based on the result of training experiments.