|
|
EHD Lubrication Film Behaviour under Rapid Change of Velocity and Load
Zimmerman, Martin ; Dzimko, Marián (referee) ; Návrat, Tomáš (referee) ; Dobeš, Petr (referee) ; Hartl, Martin (advisor)
This dissertation is focused on experimental study of lubricant film behavior under transient conditions. For real machine parts such as gears, roller bearings and cam mechanism the operating conditions such as speed, load and radius of curvature of rubbing surfaces are vary during the working cycle. Change in operating conditions may pose an increased risk of the lubricating film breakdown. Direct contact of rubbing surfaces at the time of the lubricating film breakdown causes an increase in friction and wear. The high-speed CMOS camera was used to record the dynamic changes in lubricating film thickness during the experiment. The optical interferometry method has been used for reverse reconstruction of lubricating film thickness of the recorded interferograms. It was confirmed that the operating parameters have a major impact on the formation of lubricant film and its thickness and a certain combination of boundary conditions can cause lubricating film rupture. The obtained results showed, that modified topography of the rubbing surfaces can help to increase the lubrication film thickness especially in critical phases of working cycle.
|
|
|
Electrical Activity Brain Mapping
Dobeš, Petr ; Drahanský, Martin (referee) ; Kupková, Karolína (advisor)
Electrical activity of human brain is one of the most significant signals of biological origin. In order to understand and interpret electroencephalogram (EEG signal) correctly, it is often necessary to perform its visualization. This bachelor thesis deals with EEG signal and its visualization using topographic mapping. The work includes the basics of theory and processing of EEG signal. Moreover, this work consists of design proposal and implementation of an application for topographic mapping of EEG signal obtained using Emotiv Epoc Headset device. Visualization is performed in real time (at the time of measurement). Visualized quantities are amplitude and frequency domain with the possibility to select frequency bands. Implemented application represents an alternative to the procedure when EEG signal has to be recorded and stored in order to perform its visualization.
|
|
|
Revitalization of hydraulic circuits control tester
Korsa, Filip ; Dobeš, Petr (referee) ; Nevrlý, Josef (advisor)
High-quality simulation of hydraulic mechanism saves the time and other costs associated with the changes of inappropriately designed circuit. The aim of this thesis is to revitalize the test rig intended for simulating hydraulic circuits in order to verify the results of the simulation program Hyvos used internally by Bosch Rexroth Company. It is expected that the simulation of the hydraulic machinery cycles under the influence of disturbances will be carry out on the test rig and measured data are compared with results of the virtual models which are designed using the mentioned software Hyvos. The work deals with the so called position and pressure control which represent the classical modes of pressing machines or hydraulic press brakes operation cycles. The parameters of the hydraulic closed-loop axis controller for electrohydraulic drives (HACD) are set using the software Bodac.
|
|
|
Web Tool for Estimation of Document Size
Dlouhá, Simona ; Dobeš, Petr (referee) ; Herout, Adam (advisor)
Aim of this thesis is to design and create web tool, which will allow users to get a text range in standard pages from a PDF file. The thesis introduces algorithms designed for retrieving statistics from a file. The tool is implemented using Python and Django framework, user interface is realized by HTML, CSS and JavaScript. Result of this thesis is a tool that provides text statistics, image statistics and chapters overview. Benefit of this tool is also more precise page counting in documents with images in comparison with other tools.
|
|
| |
|
|
Automatic Image Analysis for Production Quality Control of Textile
Sýkorová, Tereza ; Dobeš, Petr (referee) ; Zemčík, Pavel (advisor)
This work deals with the classification of defects that occur in the production of nonwovens. The defect classification task is part of a system for automatic production quality control. The goal is to implement a method that will classify problematic defect classes with sufficient accuracy. That was achieved using convolutional neural networks (CNN). The best results were achieved by the EfficientNet network, which had an accuracy of 81% when evaluated by cross-validation on an available dataset. Within the work, a number of experiments are performed, which are focused on the modification of input data. The influence of the shape and composition of the input images on the final classification is examined. A CNN model was also implemented, which uses additional information for classification in addition to the image.
|
|
|
Automatic Image Analysis for Production Quality Control of Textile
Sýkorová, Tereza ; Dobeš, Petr (referee) ; Zemčík, Pavel (advisor)
This work deals with the classification of defects that occur in the production of nonwovens. The defect classification task is part of a system for automatic production quality control. The goal is to implement a method that will classify problematic defect classes with sufficient accuracy. That was achieved using convolutional neural networks (CNN). The best results were achieved by the EfficientNet network, which had an accuracy of 81% when evaluated by cross-validation on an available dataset. Within the work, a number of experiments are performed, which are focused on the modification of input data. The influence of the shape and composition of the input images on the final classification is examined. A CNN model was also implemented, which uses additional information for classification in addition to the image.
|
|
|
Generative Neural Networks for Handwritten Text
Ševčík, Pavel ; Dobeš, Petr (referee) ; Hradiš, Michal (advisor)
The aim of this study was to create a generative neural network for handwritten text lines. The model produces variable-sized images of handwritten text lines based on the expected style. The proposed method exceeds existing models in the image quality and can be used to generate both individual words and entire lines of handwritten text. It combines the use of the attention mechanism to extract the features for each character from the text query and their arranging on the line by inserting spaces between them. The new approach allows more granular control of the symbol positions on the line, which leads to smoother style interpolations. In contrast to the previous approach, the proposed method uses the Gaussian filter to spread the individual symbols features to the surrounding area. This approach also allows to train the model for symbols position predictions using the adversarial loss (GAN). In addition, annotations of symbol horizontal positions on the lines of the IAM dataset of handwritten text have been created.
|
|
| |
|
| |
guest ::
RSS feed.