National Repository of Grey Literature 7 records found  Search took 0.00 seconds. 
Decoding visual stimuli from cortical activity
Vašek, Vojtěch ; Antolík, Ján (advisor) ; Šikudová, Elena (referee)
This thesis aims to develop a machine learning model that can decode stimu- lus images from cortical activity in the primary visual cortex (V1) to understand the relationship between V1 activity and visual perception. The limited avai- lability of biological data makes it necessary to use the spiking neural network model of V1 to generate the underlying training data. Machine learning tech- niques, particularly neural networks, will be explored to generate high-quality stimulus images. Standard loss functions, as well as discriminator loss from GAN networks training, will be used to train the decoding models. Linear regression models will be used baseline. The research questions to be addressed include the best decoding approach, the impact of the number of neurons recorded or stimuli presented, the loss of information in high frequencies domain and the effect of intrinsic noise in neural responses on reconstructing visual stimuli. This thesis proposes a trainable convolutional network, which outperforms other baseline models such as linear regression. We observe that the loss function producing the best results is the MSSSIM. However, the intrinsic noise in neural respon- ses limits the reconstruction, and only low frequencies are being reconstructed. The size of the dataset and the number of cortical...
Automated number plate recognition from low quality video-sequences
Vašek, Vojtěch ; Franc, Vojtěch (advisor) ; Šikudová, Elena (referee)
The commercially used automated number plate recognition (ANPR) sys- tems constitute a mature technology which relies on dedicated industrial cam- eras capable of capturing high-quality still images. In contrast, the problem of ANPR from low-quality video sequences has been so far severely under- explored. This thesis proposes a trainable convolutional neural network (CNN) with a novel architecture which can efficiently recognize number plates from low-quality videos of arbitrary length. The proposed network is experimentally shown to outperform several existing approaches dealing with video-sequences, state-of-the-art commercial ANPR system as well as the human ability to recog- nize number plates from low-resolution images. The second contribution of the thesis is a semi-automatic pipeline which was used to create a novel database containing annotated sequences of challenging low-resolution number plate im- ages. The third contribution is a novel CNN based generator of super-resolution number plate images. The generator translates the input low-resolution image into its high-quality counterpart which preserves the structure of the input and depicts the same string which was previously predicted from a video-sequence. 1
Vašek, Vojtěch ; Kučera, Petr (advisor) ; Škoda, Dominik (referee)
This work looks into using digital image processing to create an educative game, which should guide its user through interesting places in the world. Invented game named O-Hunter, which consists of two parts - server and client, uses photo comparison, analyzes daytime and uses filters to modify images. In the first section, rules of the game are described, analysis of used methods follows. Implementation is described in third section and the next section covers how the program should be used. This program shows, that some of the selected approaches are very well suitable for this application and others can fail in real situations. Last section covers conclusions and suggestions for improvements. The game O- Hunter can be employed to discover attractive places in an unusual way, its design provides option to substitute or improve implementations of used techniques. Powered by TCPDF (
Using of 3D printed patterns for mass production of castings on "green sand" moulding lines
Vašek, Vojtěch ; Lauermann, Jaroslav (referee) ; Horáček, Milan (advisor)
Shortening of the time required to make first product after demand is a crucial factor in every field of industry, not excluding the foundry. To lower this delay, there are rapid prototyping methods. The aim of this thesis is to execute an experiment and then evaluate possibilities of using patterns made by 3D FDM printer on an automatic molding line.
Order Processing Management in the Company Focusing on Business Activity
Vašek, Vojtěch ; Tesař, Rostislav (referee) ; Jurová, Marie (advisor)
The diploma thesis deals with order processing management in selected company focusing on business activity and development. The first part introduces the company and its product portfolio. The second part contains the theoretical background necessary to understand the issue. The third part focuses on analyzing the current state of the company and identifying ineffective processes within the sales department. The fourth part contains specific proposals to remove the current shortcomings and the last part summarizes the conditions and benefits of their implementation.
The Studies Order Processing in Selected Company
Vašek, Vojtěch ; Bazika, Robert (referee) ; Jurová, Marie (advisor)
Bachelor thesis deals with the study of order processing in the company HaSaM s.r.o., which focuses on developing and manufacturing of electronic payment and ticketing systems. Theoretical part focuses on explanation of the basic terms, which are related to the order and manufacturing process. The practical part contains the current process of undertaking the order and proposal for improvement in communication with suppliers and delivery terms.
Possibilities of using modern methods in manufacturing of prototype castings
Vašek, Vojtěch ; Mikulka, Vít (referee) ; Horáček, Milan (advisor)
Rapid Prototyping technologies are an important part of the industry. They allow reducing the time between the demand and the crafted prototype component. The main aim of these theses is to create a literature survey, focused on the application of these technologies in the foundry industry in the production of prototype castings.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.