|
|
Human Hand Geometric Parameters Measurement Using Numerical Methods of Image Processing
Vanžurová, Anna ; Procházková, Jana (referee) ; Štarha, Pavel (advisor)
This thesis deals with a matematic model used for estimation of the surface of a human hand. By using the theory for digital image processing, analytic geometry and numerical methods a model for proccesing a photografy of a hand was created. This model further calculates the surface in the software Matlab. The model uses tools such as median filter, adaptive tresholding or conditional erosion. Interesting is the usage of the curvature of a curve, which can be used to find the peaks and bases on the border of the hand. Dependencies between the calculated results and the measured real surfaces can be found. Therefore a regression function could be found, which gives us a point estimation of the human hand surface.
|
|
|
X-ray computed tomography of mouse embryo
Šejnohová, Marie ; Walek, Petr (referee) ; Kolář, Radim (advisor)
The aim of this semestral thesis is to compare the possibilities of available micro-CT systems. Theoretic part of this thesis there deals with possibilities of staining soft tissues and embryos because of enhancement the contrast of micro-CT images. Here follows a description of sources X-ray and detectors of available micro-CT systems. In practice there was realized the staining of embryo in cooperation with Department of histology and embryology in Brno. Then followed a measuring on FSI in Brno, ČVUT in Prague and synchrotron Elettra in Italy. In semestral thesis are described of the thesis there are compared the micro-CT systems and results of measuring embryos by means of presented systems and results are compared.The best results were obtained on micro-CT in Brno, where X-ray tube and flat panel detector were used. This images were used for a segmentation of cartilage olfactory system by means of 3D region growing. From results they were created 3D models for comparison with a manually segmented model. A less accurate results were obtain with the semi-automatic segmentation but this method isn’t too time-consuming.
|
|
|
Detection of ventricular extrasystoles
Svánovská, Zuzana ; Mézl, Martin (referee) ; Sekora, Jiří (advisor)
Ventricular extrasystoles are pathological changes in the ECG signal. Detection of ventricular extrasystoles on 12leads ECG was created in MATLAB. My work contains two algorithms. The first of these algorithms is based on comparision wides of QRS komplexes. The second algorithm matchs maximum and minimum evaluations of QRS komplexes. We look for agreements beetween these two algorithms and finally if we find these agreenments in seven leads at least we will suppose presence of ventricular extrasystoles.
|
|
|
QRS detection based on Stockwell transform
Kašík, Ondřej ; Kozumplík, Jiří (referee) ; Smital, Lukáš (advisor)
This bachelor´s thesis deals with the detection of QRS complexes in ECG record. The thesis provides a brief information related to the heart anatomy, generation of electrical signals in the heart, recording and description of the ECG record. In more detail, there is a description of the detection of QRS complexes by various methods and realization of a detector based on Stockwell transform, Shannon energy and adaptive thresholding. The evaluation process of the detection efficiency is also included. Sensitivity and positive prediction of the proposed detector on the complete MIT-BIH Arrhythmia database reached 99.80 % and 99.88 % respectively.
|
|
|
Application for Processing of a Photographed Text
Genčúr, Martin ; Rozman, Jaroslav (referee) ; Grulich, Lukáš (advisor)
This work introduces main approches for converting a photographed text into black and white form. It analyses particular methods being used for this task. Following part describes implementation of the application performing a conversion. The program is tested with suitable data (coloured picture, a picture with various shades etc.) and illustrates the usability of the application. The thesis also contains an introduction to optical character recognition (OCR) and suggests potential ways of development of this application.
|
|
|
Optical Character Recognition
Bílek, Jan ; Žák, Pavel (referee) ; Zuzaňák, Jiří (advisor)
This work describes problems of optical character recognition and methods which are used to improve its results. It specializes in problems of image spam and its adjustment for its recognition. This work describes methods of processing an image which leads to better results of text recognition and also some methods used to the character recognition. There is also described a scheme and an implementation of a system which processes an image spam and detects a text within it.
|
|
|
Recognition of Objects in Pictures
Nedoma, David ; Samek, Jan (referee) ; Zbořil, František (advisor)
This thesis is about solving a problem of recognition of objects in pictures. The aim was to create a program that will be able to recognize objects in an image. Describes progressively step by step processing of image data. Shortly describes preprocessing of image, after that describes in detail segmentation, description of segmented data and classification of objects. Describes algorithms and methods that are applicable for each step.
|
|
|
Image based road surface segmentation with lane detection
Turoň, Rudolf ; Honec, Peter (referee) ; Zemčík, Tomáš (advisor)
This bachelor thesis deals with road semantic segmentation with lane lines detection. The aim of the thesis is to conduct a survey of methods used for general segmentation and lane lines detection, collect a representative dataset for testing, design a data processing system including implementation for semantic segmentation and detection of lanes in image, video and live camera input. In preprocessing phase images are filtered using morphological transformations. The segmentation is performed using the watershed method with labels, and an adaptive algorithm is designed to find the labels. The lanes are searched for in the thresholded projectively transformed image using the sliding window technique. As a result, a segmentation accuracy of 88.3% is achieved based on the IoU metric. The thesis concludes with a discussion of the obtained results and summarizes the possibilities for further improvements of the system.
|
|
|
Application for Processing of a Photographed Text
Hotař, Roman ; Rozman, Jaroslav (referee) ; Grulich, Lukáš (advisor)
This work describes conversion methods of full-colour picture with text ( colour transition, various shades etc. ) into black and white form. It focuses mainly on adaptive thresholding. Mentions used data structures and describes interface of result application. The thesis describe at the conclusion the results of several functionality tests.
|
|
|
Application of Approximate Computing in Image Processing
Hruda, Petr ; Vašíček, Zdeněk (referee) ; Bidlo, Michal (advisor)
This master thesis focuses on approximate computing applied to image processing. Specifically, the approximation is applied to adaptive thresholding. Two approaches were used, the design of a new system using approximated components and the approximation of an existing algorithm. The resulting effect on thresholding quality was investigated. Experimental evaluation of the first approach shows quality improvements of thresholding with usage of aproximated components. Also, area of found aproximated solutions is smaller. Evaluation of the second approach shows worse quality of thresholding with usage of aproximated components. The second approach is then declared inappropriate.
|
guest ::
