|
|
The space data analysis
Burešová, Klára ; Štarha, Pavel (referee) ; Procházková, Jana (advisor)
This thesis deals with the analysis of spatial data and 3D descriptors for object recognition in scenes. The aim of the thesis was to study the issues of working with 3D data and to make a study of different types of 3D descriptors. In this work, we analyzed some descriptors in terms of their ability to describe and quantify the properties of objects in three-dimensional space. Based on this analysis, we selected a suitable descriptor for object recognition in scenes and implemented it. The result of this work is a deeper understanding of 3D descriptors and their applications in spatial data analysis. Our implementation of the selected descriptor provides a basis for further research and development in object recognition in 3D scenes.
|
|
|
Material Decomposition in Spectral Submicron Computed Tomography
Mikuláček, Pavel ; Zikmund, Tomáš (referee) ; Štarha, Pavel (advisor)
Počítačová tomografie je nedestruktivní zobrazovací metoda vnitřních struktur vzorků. Využití nachází především v lékařství, ale také v průmyslu nebo výzkumu. V lékařství se také běžně používá spektrální počítačová tomografie, která umožňuje diferenciaci materiálů na základě jejich útlumových vlastností, které jsou závislé na energii rentgenového záření. Tento přístup se začíná také objevovat v průmyslových aplikacích. Tato práce se zabývá dvouenergetickou počítačovou tomografií, což je specifická forma spektrální tomografie využívající dvě různá rentgenová spektra. Dekompozici materiálu lze provádět buď na projekčních datech nebo na jejich rekonstrukcích. Výstupem jsou obrazy obsahující informace o koncetracích jednotlivých materiálů, které se nachází ve skenovaném objemu. Kromě koncetrací lze také zjistit atomová čísla a hustoty těchto materiálů. V této práci byla implementována základní metoda, jejíž výstupy slouží jako počáteční odhady v řadě iteračních algoritmů, které provádí dekompozici. Také byla provedena implementace metody, která redukuje šum v rozložených obrazech, jejichž hodnoty jsou vysoce korelované. Algoritmy byly testovány na umělých datech, ale také na reálných datech pořízených přístrojem Rigaku Nano3DX. Bylo provedeno srovnání vlivu různých tomografických artefaktů na výsledný rozklad materiálu
|
|
|
Alternative JPEG image decoder
Bureš, Jiří ; Štarha, Pavel (referee) ; Rajmic, Pavel (advisor)
This thesis deals with the JPEG image codec, edge detection in images, sparse signal representations and proximal algorithms. First, the operation of the JPEG encoder and decoder and the theory underlying it are described. Then, based on the theoretical knowledge, a new proximal algorithm is constructed and implemented in an existing JPEG algorithm in order to remove block relics in the decoded image. The programming side is solved in Matlab environment. The results are evaluated using MSE, PSNR and SSIM methods.
|
|
| |
|
|
Image Processing for Optimal Focusing of Light Microscope
Kvašňovská, Petra ; Procházková, Jana (referee) ; Štarha, Pavel (advisor)
The aim of this bachelor thesis is a selection of appropriate parameters to gauge image sharpness. The thesis contains the mathematical theory needed to solve this issue. Picture digitalization necessary for working with an image on a computer is also described. The outcome of the thesis is a method, using which the positions of optimally sharpened test images of objects observed by a light microscope were found.
|
|
|
Pattern recognition by means of moment method
Števček, Juraj ; Štarha, Pavel (referee) ; Druckmüller, Miloslav (advisor)
This diploma thesis deals with object recognition using moment method. We will recognize the letters from the scanned text. The moment method works with principal moments, these are moments invariant to scaling, translation and rotation. This thesis focuses on a detail description of the moment method, on determining the optimal combination of principal moments suitable for object recognition and to create a program for recognition. For all combinations of principal moments the accuracy of the recognition is calculated and the best option is proposed. The moment method will be compared with the moment invariants method. This method is similar to the moment method but has a different approach to rotation invariants.
|
|
|
The methods of texture mapping on point cloud
Květný, Michal ; Štarha, Pavel (referee) ; Procházková, Jana (advisor)
This diploma thesis deals with a mapping of textures obtained from ordinary photographs onto the three-dimensional point clouds created by a 3D scanner device. The thesis proposes a mapping method that consists of reconstruction of a sparse point cloud from input photos and registration of the point cloud with an input dense point cloud using FPFH (Fast Point Feature Histogram) descriptors. The thesis also contains an implementation of the proposed method in C++ programming language which is tested on a real world data.
|
|
|
Spectral Analysis in Microscopy
Sýs, Michal ; Procházková, Jana (referee) ; Štarha, Pavel (advisor)
The aim of this work is to develop a mathematical apparatus for the accurate description of the colour of objects in a microscope image. This problem is solved by extracting the brightness intensities of pollen grains at fourteen wavelength bands and representing them in color space. The developed apparatus has been successfully applied to two series of images, allowing comparison of the color of individual objects and simplifying its description. The results obtained allow accurate investigation of the colour of objects in the image and form a simple tool for spectral analysis.
|
|
| |
|
|
Numerical Methods of Image Analysis for Optimal Lens Focusing
Mikuláček, Pavel ; Loučka, Pavel (referee) ; Štarha, Pavel (advisor)
The bachelor thesis deals with determining the optimal focus of the lens. It includes part of the mathematical theory associated with this problem. It explains the mathematical procedure that can be used to decide on the optimal focus of the lens. The result of the work is a single-purpose application that processes images of the Sun and can determine the optimal position of the focusing ring.
|
guest ::
