|
|
Interactive web applications for support of teaching signal processing
Mašek, Petr ; Rajmic, Pavel (referee) ; Záviška, Pavel (advisor)
The bachelor thesis deals with the topic of processing one-dimensional signals. Five interactive web applications have been created in JavaScript to support teaching. These are applets that relate to the following topics: rational number resampling of the discrete signal, mean and root mean square signal values, discrete linear and circular convolution in 1D, and discrete cross correlation. The thesis introduces each topic by a theoretical foreword, followed by description of implementation and user interface of the created applications and further there is a commentary on the development and testing of applets.
|
|
| |
|
|
Methods for increasing bit-depth in images
Záviška, Pavel ; Koldovský, Zbyněk (referee) ; Rajmic, Pavel (advisor)
The goal of this bachelor thesis is the application of an issue of sparse representations on the task of increasing bit-depth in images. Standard methods for bit-depth expansion are described, subsequently a method using sparse representation of an image signal is introduced. Methods are programmed in the Matlab enviroment. The results of the methods implemented are compared by using PSNR, SSIM objective indexes and subjectively using an online questionnaire.
|
|
|
Clean photo out of corrupted videosequence
Berky, Martin ; Záviška, Pavel (referee) ; Rajmic, Pavel (advisor)
This diploma thesis deals with separation of moving objects from static unchanging background in video sequence. Thesis contains description of common method of separation and approach based sparse signal representation. In the practical part of thesis, there were created video sequences, which are used to verify designed algorithm implemented in Matlab interface, disegned to obtain separated background from damaged video frames.
|
|
|
Interactive web applications for teaching purposes in the field of signal processing
Kuba, Michael ; Rajmic, Pavel (referee) ; Záviška, Pavel (advisor)
This thesis deals with selected topics of signal processing. The aim of the project is the implementation of four interactive web applications in JavaScript, intended for teaching purposes. These applications deal with following topics: rational number resampling of the discrete signal, mean and root mean square signal values, discrete linear and circular convolution in 1D, and discrete cross-correlation. The thesis is divided into the theoretical and the practical part. There is a theroretical foreword of topics the applications deal with, including an introduction to signals followed by a description of functionality and the user interface of created applications.
|
|
|
Tool for subjective quality evaluation of audiosignals restoration
Černý, Viktor ; Dorazil, Jan (referee) ; Záviška, Pavel (advisor)
Bacherol thesis deals with the theory of audio signal depreciation and reconstruction, with objective tests and subjective methods of reconstructed signal evaluation. At first, there are described several types of depreciation of audio signals and then reconstruction types for their reconstruction. After the reconstruction, types six objective and six subjective tests are mentioned for the evaluation of audio signal quality. The next part of the Bachelor thesis is the description of the program that was implemented in the programming language Java. The program is designed to design MOS, MUSHRA and ABX listening tests as well as to use for the actual testing. At the end of the thesis are the results of the listening tests that were used to test the program and the user ratings of the application.
|
|
|
Innovation of digital image processing exercises
Honkyš, Jan ; Záviška, Pavel (referee) ; Rajmic, Pavel (advisor)
The bachelor thesis deals with innovation of lectures in the field of image processing. The innovation consists in the preparation of instructions for individual exercises in alternative programs, Adobe Photoshop CS6 Extended is now used in exercises. The first chapter deals with color models, spaces, color management and bitmap file formats. The second chapter introduces alternative bitmap editors – Adobe Photoshop CC 2021, GIMP, Affinity Photo, Zoner Photo Studio X, Corel PaintShop Pro 2022. Then they are compared according to several criteria, the cost for the school and students is considered and a selection of programs suitable for teaching image processing is made. In the next chapters, instructions for individual exercises are made in selected programs and an evaluation of the suitability of these programs for teaching is performed.
|
|
|
Search for a dictionary for audiosignals
Martinek, Václav ; Záviška, Pavel (referee) ; Rajmic, Pavel (advisor)
This thesis should answer what constitutes audiosignal. It deals with the dictionary learning based on sparse representations of signals. This thesis describing algorithms, which are important for the creation of learning dictionaries. We find a comparison of signal representation using Fourier transform and learned vocabulary. It describes the creation of a database of musical sounds.
|
|
|
Audio Signal Declipping and Dequantization Using Sparsity-Based Methods
Záviška, Pavel ; Šroubek,, Filip (referee) ; Koldovský,, Zbyněk (referee) ; Rajmic, Pavel (advisor)
Audio signály jsou náchylné k různým typům poškození, přičemž jedním z nejčastějších a nejproblematičtějších druhů poškození je clipping. Tato dizertační práce se zaměřuje na rekonstrukci zvukových signálů poškozených nelineárním zkreslením s hlavním zaměřením na declipping a dekvantizaci a popisuje vědecký přínos v této oblasti s využitím metod založených na řídké reprezentaci. První část dizertační práce se zabývá problematikou declippingu a představuje několik přístupů založených na řídkých reprezentacích signálů. Součástí je jak originální výzkum, tak i převzaté algoritmy, které však byly v rámci této práce reimplementovány nebo modifikovány. Kvalita výstupů rekonstrukčních algoritmů je vyhodnocena jak pomocí ukazatele SDR, tak i s využitím percepčně založených metrik. V další části se práce zaměřuje na zakomponování psychoakustiky do problému declippingu pomocí váhování transformačních koeficientů s třemi navrženými způsoby konstrukce vah. Je zde dokázáno, že při správně zvolených vahách je možné výrazně zlepšit kvalitu rekonstrukce a vyrovnat se tak nejlepším algoritmům při zachování nízké výpočetní náročnosti. V poslední části práce je pozornost je také věnována metodám umožňujícím odchylku ve spolehlivé části signálu. V tomto směru práce zkoumá percepční vliv prostého nahrazení těchto spolehlivých vzorků, identifikuje jeho hlavní nevýhody a následně představuje metody, které kompenzují negativní efekty způsobené tímto nahrazením. Je ukázáno, že s využitím těchto technik je možné bez významného navýšení výpočetní náročnosti výrazně zlepšit dosaženou kvalitu rekonstrukce. Vybrané algoritmy jsou rovněž aplikovány na problém audio dekvantizace. Součástí práce jsou repozitáře obsahující implementace všech představených metod.
|
|
|
Modern audio denoising with utilization of phase information
Skyva, Pavel ; Záviška, Pavel (referee) ; Rajmic, Pavel (advisor)
The thesis deals with modern methods of audio denoising. Reconstruction of the audiosignal is primarly based on utilization of phase information of signals and phase derivatives. Denoising methods also use sparse signal representations. In thesis is described the way of searching sparse coefficients using proximal Condat algorithm and following computation of reconstructed signal using this coefficients. The reconstruction algorithms are implemented in the MATLAB software with toolbox LTFAT included. Results of the reconstruction are compared using objective evaluation method Signal-to-Noise Ratio (SNR) and also by subjective evaluation.
|
guest ::
RSS feed.