|
|
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.
|
|
|
Restoration of degraded audiosignals using sparse representations
Mokrý, Ondřej ; Veselý, Vítězslav (referee) ; Rajmic, Pavel (advisor)
This bachelor thesis is focused on the problem of inpainting a segment of missing samples in an audiosignal. The signal is represented as sparse vector using discrete Gabor transform. The problem of inpainting missing samples while preserving the sparsity of the representation is formulated as an optimisation task, which is then solved using Douglas-Rachford algorithm. In contrast with the state-of-the-art approaches, the algorithm is extended by proposing method for compensating the energy decrease which occurs in the restored signal.
|
|
|
Modern audio signal restoration methods
Kalník, Jan ; Mangová, Marie (referee) ; Rajmic, Pavel (advisor)
This bachelor thesis deals with modern methods of audio signal inpainting. Main methods used will be sparse signal representations. Those will be implemented into MATLAB software using LTFAT and UNLocBoX toolboxes. Next we would like to accomplish the best results with the audio inpainting and then compare these results as well as subjectively as objectively.
|
|
|
Dictionary learning for sparse signal reconstruction
Ozdobinski, Roman ; Rajmic, Pavel (referee) ; Mach, Václav (advisor)
This bachelor thesis discusses the dictionary learning for the reconstruction of signal based on sparse representations. There are methods of static and optimized dictionary of matrices described that are used with approximative Orthogonal Matching Pursuit algorithm to reconstruct the missing groups of samples in the audio signal. There is theoretically analyzed algorithm for learning K-SVD dictionary together with its implementation in Matlab. Furthermore, the selected dictionaries are compared to the various types of audio signals.
|
|
|
Methods of acquisition and processing of images based on sparse representations
Talár, Ondřej ; Mach, Václav (referee) ; Rajmic, Pavel (advisor)
Thesis deals with the reconstruction possibilities provided by the sparse representation of signals. This representation reduces the signal to a mere vector of elements which indicate the signal portion in the dictionary array. It outlined the problems with the quantized signal and recalled modulation type, involving a quantization and its ways. The solution is selected Douglas-Rachford algorithm that allows us to approximate on to the set of all acceptable solutions. At the end is demonstrated problem solution and several tests for presentation of created program.
|
|
|
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.
|
|
|
Inpainting of Missing Audio Signal Samples
Mach, Václav ; Polec,, Jaroslav (referee) ; Koldovský,, Zdeněk (referee) ; Rajmic, Pavel (advisor)
V oblasti zpracování signálů se v současné době čím dál více využívají tzv. řídké reprezentace signálů, tzn. že daný signál je možné vyjádřit přesně či velmi dobře aproximovat lineární kombinací velmi malého počtu vektorů ze zvoleného reprezentačního systému. Tato práce se zabývá využitím řídkých reprezentací pro rekonstrukci poškozených zvukových záznamů, ať už historických nebo nově vzniklých. Především historické zvukové nahrávky trpí zarušením jako praskání nebo šum. Krátkodobé poškození zvukových nahrávek bylo doposud řešeno interpolačními technikami, zejména pomocí autoregresního modelování. V nedávné době byl představen algoritmus s názvem Audio Inpainting, který řeší doplňování chybějících vzorků ve zvukovém signálu pomocí řídkých reprezentací. Zmíněný algoritmus využívá tzv. hladové algoritmy pro řešení optimalizačních úloh. Cílem této práce je porovnání dosavadních interpolačních metod s technikou Audio Inpaintingu. Navíc, k řešení optimalizačních úloh jsou využívány algoritmy založené na l1-relaxaci, a to jak ve formě analyzujícího, tak i syntetizujícího modelu. Především se jedná o proximální algoritmy. Tyto algoritmy pracují jak s jednotlivými koeficienty samostatně, tak s koeficienty v závislosti na jejich okolí, tzv. strukturovaná řídkost. Strukturovaná řídkost je dále využita taky pro odšumování zvukových nahrávek. Jednotlivé algoritmy jsou v praktické části zhodnoceny z hlediska nastavení parametrů pro optimální poměr rekonstrukce vs. výpočetní čas. Všechny algoritmy popsané v práci jsou na praktických příkladech porovnány pomocí objektivních metod odstupu signálu od šumu (SNR) a PEMO-Q. Na závěr je úspěšnost rekonstrukce poškozených zvukových signálů vyhodnocena.
|
|
|
Modern methods for restoration of degraded audiosignals
Mokrý, Ondřej ; Koldovský,, Zbyněk (referee) ; Rajmic, Pavel (advisor)
The master's thesis deals with the problem of restoring a block of missing samples in a digital audio signal. This problem is formulated as an optimization task, which seeks the sparsest time-frequency representation of a signal within the set of feasible reconstructed signals. Several particular formulations are discussed, namely the analyzing and the synthesizing model, both for convex and non-convex approaches. Suitable algorithms are proposed for solving these formulations, and in the convex case, the method is further enhanced by various procedures to compensate for the energy drop in the inpainted signal segment. The proposed algorithms are tested on real recordings, and their performance is shown to be competitive with the state-of-the-art.
|
|
|
Separation of dynamic and static structures in a series of images
Gebrtová, Karolína ; Kosová, Petra (referee) ; Rajmic, Pavel (advisor)
This bachelor thesis is focused on the separation of static and dynamic structures of video. While the major part of the video background remains static, a few moving particles can be observed. Such video can be represented by a low-rank and a sparse component. Using a low-rank structure, the background can be separated by applying either the median filter or the dynamic mode decomposition. Furthermore, the thesis includes the robust principal component analysis which can be perceived as an optimization problem. In addition, the thesis is enhanced by result visualization methods.
|
|
|
Restoration of signals after passing through the limiter with the use of psychoacoustic model
Kramář, Denis ; Rajmic, Pavel (referee) ; Záviška, Pavel (advisor)
This bachelor thesis deals with the use of sparse representaions for the purpose of restoration clipping-damaged audiosignal. First, a theory of limiter and signal limiting itself is discussed. Subsequently, some of present methods based on sparse representations theory are given. The theory of sparse representations is discussed in following chapture. After that is here described a psychoacoustic model and it's use for declipping. At the end of theoretical part, two methods dealing with this problem are introduced. First is based on synthesis model of signal using Douglas-Rachford algorithm. Second is based on analysis signal model using Chambolle-Pock algorithm. In the next part is their implementation in the Matlab environment. Finally, the result achieved by both algorithms are evaluated.
|
guest ::
