National Repository of Grey Literature 166 records found  beginprevious90 - 99nextend  jump to record: Search took 0.00 seconds. 
Automatic recognition of meaning in texts
Jeleček, Jiří ; Dvořák, Pavel (referee) ; Povoda, Lukáš (advisor)
As part of this work it was designed and implemented a system using data mining techniques from the text in order to detect emotions in Czech, English and German language texts. Because the system is built mostly on machine learning techniques, was designed and created training set, which was later used to build the model classifier using the selected algorithms.
Application for a Demonstration of the Histogram of Oriented Gradients Method for Object Detection
Mrázek, Zdeněk ; Dvořák, Pavel (referee) ; Říha, Kamil (advisor)
The target of this thesis is summarize the theory of method Histogram of oriented gradients and process algorithm for demonstration and visualization HOG descriptor, train SVM algorithm and subsequent detection of the object. For the work environment was selected MS Visual Studio 2012 using the object-oriented C++ language with using OpenCV library.
Segmentation of MR images using machine learning algorithms
Dorazil, Jan ; Mikulka, Jan (referee) ; Dvořák, Pavel (advisor)
This thesis concerns with magnetic resonance image segmentation using Random Forests algorithm. Employed technologies accomplishing the specified task include C++ progra- mming language with libraries ITK and OpenCV. This work descibes the technique of processing images from loading through preprocessing to the actual segmentation. The outcome from this work is a programme that automatically segmentates MR images of mouse’s head to the brain and the surroundings.
Database of non-biometric and soft biometric human traits
Mlýnková, Barbora ; Říha, Kamil (referee) ; Dvořák, Pavel (advisor)
This paper is dealing with the detection of secondary biometric and nonbiometric characteristics in the picture, for the use of person identification. These characteristics are described and categorized. It also deals with the design of the database structure, which works with masks and categories of characteristics for their processing. This work might be practically used for machine learning.
Correction of artifacts in diffusion-weighted MR images
Petrek, Tomáš ; Dvořák, Pavel (referee) ; Marcoň, Petr (advisor)
The semester project is described by nuclear magnetic resonance, is discussed further in diffusion MRI and its method of measurement. The next section discusses the creation and correction of artifacts in MRI, and in the last part of the experiment made ??various filtering image distortion artifacts.
Image fusion in thermal spectrum
Petrásek, Daniel ; Dvořák, Pavel (referee) ; Mekyska, Jiří (advisor)
In this paper, there is mentioned necessary theory of image processing and fusion, which belongs to a group of the most used ways of image processing in the present. In the first part there is mentioned basic findings of image and electromagnetic spectrum. In the second part there is discussed some facts of infrared radiation, thermal cameras, then it is continued with elementary methods of image focus evaluation, segmentation and noise thresholding. In next part there is introduced scheme of image fusion system and basic idea of its implementation. In the end of this thesis there is described implemented system of image fusion, detailed description of reached results, thesis rating and few ideas of improving whole system.
Study of System with RC Model Servos
Dvořák, Pavel ; Maršík, Lukáš (referee) ; Zemčík, Pavel (advisor)
The goal of this bachelor's thesis was to investigate how works modeller servos and how they can be controlled by a computer. In this thesis was created movable stand for camera, intended for installation in aircraft model. The servos are controlled by Arduino board.
Chess Program with Various Chess Variations with Various Initial Arrangement
Dvořák, Pavel ; Žák, Jakub (referee) ; Rozman, Jaroslav (advisor)
This thesis focuses on describing required components in the process of creation of modern chess application. Goal is to create chess program with several chess variations using structures and algorithms based on professional chess programs. Thesis describes principles of chessboard representation and various factors of chessboard state evaluation used in classic chess and implemented variants. Finally thesis describes game-tree search algorithms and enhancements and compares their effect.
Content Based Photo Search
Dvořák, Pavel ; Beran, Vítězslav (referee) ; Španěl, Michal (advisor)
This thesis covers design and practical realization of a tool for quick search in large image databases, containing from tens to hundreds of thousands photos, based on image similarity. The proposed technique uses various methods of descriptor extraction, creation of Bag of Words dictionaries and methods of storing image data in PostgreSQL database. Further, experiments with the implemented software were carried out to evaluate the search time effectivity and scaling possibilities of the design solution.
Brain Tumor Detection and Segmentation in Multisequence MRI
Dvořák, Pavel ; Přibil, Jiří (referee) ; Šprláková-Puková,, Andrea (referee) ; Smékal, Zdeněk (advisor)
Tato práce se zabývá detekcí a segmentací mozkového nádoru v multisekvenčních MR obrazech se zaměřením na gliomy vysokého a nízkého stupně malignity. Jsou zde pro tento účel navrženy tři metody. První metoda se zabývá detekcí prezence částí mozkového nádoru v axiálních a koronárních řezech. Jedná se o algoritmus založený na analýze symetrie při různých rozlišeních obrazu, který byl otestován na T1, T2, T1C a FLAIR obrazech. Druhá metoda se zabývá extrakcí oblasti celého mozkového nádoru, zahrnující oblast jádra tumoru a edému, ve FLAIR a T2 obrazech. Metoda je schopna extrahovat mozkový nádor z 2D i 3D obrazů. Je zde opět využita analýza symetrie, která je následována automatickým stanovením intenzitního prahu z nejvíce asymetrických částí. Třetí metoda je založena na predikci lokální struktury a je schopna segmentovat celou oblast nádoru, jeho jádro i jeho aktivní část. Metoda využívá faktu, že většina lékařských obrazů vykazuje vysokou podobnost intenzit sousedních pixelů a silnou korelaci mezi intenzitami v různých obrazových modalitách. Jedním ze způsobů, jak s touto korelací pracovat a používat ji, je využití lokálních obrazových polí. Podobná korelace existuje také mezi sousedními pixely v anotaci obrazu. Tento příznak byl využit v predikci lokální struktury při lokální anotaci polí. Jako klasifikační algoritmus je v této metodě použita konvoluční neuronová síť vzhledem k její známe schopnosti zacházet s korelací mezi příznaky. Všechny tři metody byly otestovány na veřejné databázi 254 multisekvenčních MR obrazech a byla dosáhnuta přesnost srovnatelná s nejmodernějšími metodami v mnohem kratším výpočetním čase (v řádu sekund při použitý CPU), což poskytuje možnost manuálních úprav při interaktivní segmetaci.

National Repository of Grey Literature : 166 records found   beginprevious90 - 99nextend  jump to record:
See also: similar author names
86 DVOŘÁK, Pavel
130 DVOŘÁK, Petr
5 Dvořák, P.
130 Dvořák, Petr
2 Dvořák, Petr,
1 Dvořák, Prokop
1 Dvořák, Přemysl
Interested in being notified about new results for this query?
Subscribe to the RSS feed.