|
Sběr a zpracování dat při měření LED modulů předních světlometů
Polák, Petr ; Honzík, Petr (referee) ; Holek, Radovan (advisor)
The aim of this is devise a file of measured parameters for LED Module Check device which is using for measurement of headlamps LED modules and for following generating reports of measurements. At the beginning of the thesis is described LED Module Check functionality and LED parameters to following up. In the following section is decribed main class written in C# language, own measured values and their interpretation to XML format. Finally, are introduced tools for making and adjustment documents and conversion to PDF format.
|
|
Dynamic Time Warping for Vehicle Classification
Halachkin, Aliaksei ; Honec, Peter (referee) ; Honzík, Petr (advisor)
This thesis focuses on the dynamic time warping. During the work was written C/Python library. Using this library, the algorithm was subsequently applied for the vehicle classification based on their shapes. Testing had been performed on real data from a laser scanner. Then the algorithm had been compared to the correlation and Euclidean distance. Finally, laboratory model had been created, which demonstrates vehicle recognition using dynamic time warping.
|
|
Plagiarism Detection
Kobath, Martin ; Petyovský, Petr (referee) ; Honzík, Petr (advisor)
The Bachelor‘s thesis concentrates on plagiarism detection in written text, mainly final theses in author’s Alma mater. The current situation of automatic plagiarism detection of text in czech language is evaluated and a theoretic analysis directed on classic detection by finding matches in a text with use of external sources, mainly the internet, is made. Project ends with custom concept of automatic plagiarism detection system and with an experiment made on 100 randomly chosen theses from The Faculty of Electrical Engineering and Communication.
|
| |
|
Data analysis from the manufacturing process
Krčmář, Martin ; Honzík, Petr (referee) ; Zezulka, František (advisor)
This thesis deals with the classification of production data using algorithms: neural networks, decision trees and naive bayesian classifier. The neural network is dedicated forward multilayer networks with a learning algorithm of backpropagation. In thesis, these algorithms are described and evaluated their pros and cons. Another part deals with the development of the program in C# for creating these algorithms. The last part is devoted to the evaluation of the results. Bachelor thesis contains a sample of generated clasification models decision tree and bayesian classifier.
|
| |
|
Feature Selection Based on Combination of Uncorrelated Evaluation Functions
Vaculík, Karel ; Klusáček, Jan (referee) ; Honzík, Petr (advisor)
In order to process large amount of data, it is necessary to use computers. It is possible to use statistical methods or machine learning in some cases. In either case, data can be represented with large number of features. Selection of suitable subset of features can be crucial for efficient processing. This thesis explores a subgroup of feature selection methods which are called filter methods. Comparison of such methods is carried out and the results are used in the design of a new method. This new method uses a combination of existing methods.
|
| |
|
Pre-Amplifier for MEMS Microphone
Ryšavý, Jindřich ; Honzík, Petr (referee) ; Schimmel, Jiří (advisor)
Thesis discusses the possibility of using MEMS microphones in measuring systems. Describes the characteristics of MEMS components and shows possible realization of analog to digital signal convertor when a microphone with analog output is used. Design of the amplifier is made with respect to low noise and low power consumption. Also is shown the possibility of using antialliasing filter as microphone frequency response correction at the same time.
|
| |