|
|
Methods for respiration estimates from ECG signal
Mitrengová, Jana ; Mézl, Martin (referee) ; Králík, Martin (advisor)
The thesis deals with the realization of methods for estimation of the respiratory curve from the ECG signal. The first part of the thesis deals with the anatomy and physiology of the respiratory and cardiovascular system. In this part of the thesis are also described ways of the breathing monitoring. The second part of the thesis is dedicated to the description of individual methods for the ECG derived respiration. The third part deals with the realization of selected methods, application of method algorithms on real data and comparison of resulting respiratory curves with the respiratory signals available from the PhysioNet database. In conclusion, the individual methods are compared with each other.
|
|
|
Separation of dynamic and static structures in a videosequence
Serediuk, Vadym ; Pavlík, Jan (referee) ; Rajmic, Pavel (advisor)
The aim of this bachelor thesis is the description and implementation of methods for the separation of low-rank and sparse structures in image sequence, mainly focusing on methods based on convex optimization. The beginning of the work is devoted to the description of basic mathematical concepts and definitions used in the work. The next comes the description of general case of background-foreground separation in video, followed by a brief overview of existing approaches based on various mathematical models. Then, methods based on matrix decomposition are described in more detail, a convex minimization task for background separation is defined. This is followed by a practical part, where the used methods are first described and then implemented in MATLAB scripts. At the end, the analysis and comparison of the results (obtained both using own scripts and external library) is performed.
|
|
|
QRS complex detection in multilead ECG signals
Šlancar, Matěj ; Smital, Lukáš (referee) ; Kozumplík, Jiří (advisor)
The aim of this thesis is to introduce the principles of software QRS detection, which is based on different combinations of orthogonal (pseudoorthogonal) leads. The thesis describes the main components of the ECG signal, a selection of methods that can be used for QRS detection in orthogonal leads and finally the evaluation of the effectiveness of the chosen methods and a comparing the approaches with the results of other authors. Functionality of detection algorithm has been tested on signals of CSE standard library.
|
|
|
Multispectral Image Processing
Li, You ; Juránek, Roman (referee) ; Zemčík, Pavel (advisor)
S rychlým rozvojem technologie multispektrálního zobrazování v posledních desetiletích obrázky získané zobrazovacími systémy obsahují nejen barevná pásma RGB v každodenním životě, ale také mají multispektrální barevná pásma a vysoké prostorové rozlišení v multispektrálních obrazových datech. Díky tomu obrázky obsahují bohaté informace o charakteristických cílových oblastech. Fúze obrazu je také důležitou větví v oblasti zpracování obrazu, kde je více obrázků ze stejné oblasti ve stejné výšce sloučeno do jednoho obrazu. Poté se zlepší korelace mezi spektrálními informacemi multispektrálních obrazů. Aby se informace na obrázku neztratily. Tato práce obsahuje popis návrhu a implementace multispektrálního obrazového systému, předzpracování multispektrálních obrazů, fúzi multispektrálních obrazů a analýzu hlavních komponent. Nakonec je představeno hodnocení celého systému.
|
|
|
Human body detection in a video scene
Šmirg, Ondřej ; Číka, Petr (referee) ; Kohoutek, Michal (advisor)
The project consists of two distinct levels i.e. separation level and diagnostic level. At the separation level, statistical models of gaussians and color are separately used to classify each pixel as belonging to backgroung or foreground. Adopted method is mixture of gaussians.A mixture of gaussians model is suitable here because the results of the picture tests will not depend on the lens opening, but rather on the colors in the backgroung. A mixture of gaussians model for return data seems reasonable. The achieved results the used method on the real sequences are presented in the thesis. Diagnostic level is identified human body on the scene. Adopted method is ASM(Active Shape Models) with PCA(Principal Component Analysis). ASM are statistical models of the shape of human bodies which iteratively deform to fit to an example of the object in a new image.
|
|
|
Analysing Tool for Generating of Drum Triggers from Downmix Record
Konzal, Jan ; Mucha, Ján (referee) ; Přikryl, Lubor (advisor)
This thesis deals with the design and implementation of a tool for generating drums triggers from a downmix record. The work describes the preprocessing of the input audio signal and methods for the classification of strokes. The drum classification is based on the similarity of the signals in the frequency domain. Principal component analysis (PCA) was used to reduce the number of dimensions and to find the characteristic properties of the input data. The method support vector machine (SVM) was used to classify the data into individual classes representing parts of the drum kit. The software was programmed in Matlab. The classification model was trained on a set of 728 drum samples for seven categories (kick, snare, hi-hat, crash, ride, kick + hi-hat, snare + hi-hat). The success of the system in the classification is 75 %.
|
|
| |
|
|
Real-time Facial Feature Tracking
Peloušek, Jan ; Mekyska, Jiří (referee) ; Přinosil, Jiří (advisor)
This thesis considers the problematic of the object recognition in a digital picture, particularly about the human face recognition and its components. There are described the basics of the computer vision, the object detector Viola-Jones, its computer realization with help of the OpenCV libraries and the test results. This thesis also describes the accurate system of the facial features detection per the algorithm of the Active Shape Models and also related mechanism of the classifier training, including the software implementation.
|
|
|
Sensory evaluation of different types of meat products
Lanžhotská, Aneta ; Vespalcová, Milena (referee) ; Diviš, Pavel (advisor)
The diploma thesis deals with sensory evaluation of selected types of meat products, specifically sausages. Different types of sausages, which contained added chemicals and spices, were compared. The theoretical part describes the properties of meat and meat products and the basic technological procedures used for their processing and production. Furthermore, added substances and spices, which are often considered as flavorings or preservatives, were characterized. The principles of correct performance of sensory analysis and sensory perception of food were also presented, which include a description of the sensory workplace, sample preparation, its implementation, procedure and various evaluation methods used. The experimental part describes the specific work tools used and the conditions under which the evaluation took place. A total of 12 types of sausages were evaluated, 7 of which came from meat production and 5 were prepared in the laboratory of food chemistry at Brno University of Technology. The differences between the samples of sausages, which differed in content and type of added chemicals and spices, were clearly shown using ray graphs. Then the Grubbs test was used, which excluded outliers from the final evaluation. These values were further excluded from further evaluation. The Kruskall-Wallis test was used to distribute the resulting mean values and to determine whether there was a statistically significant difference in sensory evaluation between the results. An appendix to the Kruskall-Wallis test Dunn's test was used to divide the resulting values into given groups according to statistically significant difference and similarity. The analysis of the main components of the so-called PCA was used to find the differences and similarities of the samples included in the groups.
|
|
|
Neural Network Based Face Localization
Hendrych, Pavel ; Šiler, Ondřej (referee) ; Švub, Miroslav (advisor)
This thesis issues with possible methods for face detection and localization according to the state of the art. It describes various approaches and it is aimed at localization by neural networks and at necessary operations that have to be done before localization and after that for correct results representation. This project contains implementation of few approaches to neural netwok based face localization with emphasis on eigenfaces based face localization as well as implementation of simple classifier using distance of reconstructed face to the original one. Detailed description of implemented system, achieved results and dependecy of system performance on it's inner settings is also provided.
|
guest ::
