|
|
Optic disc segmentation in retinal images
Vymazal, Bronislav ; Odstrčilík, Jan (referee) ; Kolář, Radim (advisor)
Cílem této práce bylo vytvořit kompaktní algoritmus sloužící k lokalizaci, segmentaci a sledování optického disku v sekvencích obrazů sítnice. Pro účely lokalizace byla použita nová metoda založena na „fast radial transform“. Segmentace optického disku je založena na klasické metodě Chan-Vese, která byla rozšířena o další prvky a implementována v lokalizované podobě, která umožňuje lépe segmentovat obrazy s vysokou mírou nehomogenity. Pro sledování optického disku v sekvencích obrazů byl využit optický tok založený na Lucas-Kanade algoritmu. Všechny metody byly vyhodnoceny na dvou databázích obrazů s odlišnými charakteristikami. První databází je klasická databáze s vysokým rozlišením. Druhou je pak databáze s nízkým rozlišením a vysokou mírou šumu. Všechny výsledky jsou pak porovnány s referenčními hodnotami a vyhodnoceny.
|
|
|
Analysis of retinal vessel pulsation
Holásková, Anna ; Mézl, Martin (referee) ; Odstrčilík, Jan (advisor)
The content of this work is the analysis of retinal vessels pulsation of video sequences acquired by experimental fundus camera based on measuring the brightness profile of the vessel. The first level of analysis in this work is the segmentation of blood vessels and diameter measurement of blood vessels during the sequence. The work contains research methods dealing with the diameter measurement and evaluation of pulsation and analysis segmentation methods using for analysing the vasculature of the retina. From these methods, a vessel tracking method was selected. In segmented video sequences is on the ground of intensity profiles analysed vessel pulsation. Analysis is also made on original dataset and results are discussed considering the frequency characteristic.
|
|
|
A Classification Methods for Retinal Nerve Fibre Layer Analysis
Zapletal, Petr ; Kolář, Radim (referee) ; Odstrčilík, Jan (advisor)
This thesis is deal with classification for retinal nerve fibre layer. Texture features from six texture analysis methods are used for classification. All methods calculate feature vector from inputs images. This feature vector is characterized for every cluster (class). Classification is realized by three supervised learning algorithms and one unsupervised learning algorithm. The first testing algorithm is called Ho-Kashyap. The next is Bayess classifier NDDF (Normal Density Discriminant Function). The third is the Nearest Neighbor algorithm k-NN and the last tested classifier is algorithm K-means, which belongs to clustering. For better compactness of this thesis, three methods for selection of training patterns in supervised learning algorithms are implemented. The methods are based on Repeated Random Subsampling Cross Validation, K-Fold Cross Validation and Leave One Out Cross Validation algorithms. All algorithms are quantitatively compared in the sense of classication error evaluation.
|
|
|
Texture Analysis of Ophthalmologic Images
Kaňka, Jan ; Odstrčilík, Jan (referee) ; Jan, Jiří (advisor)
This thesis is concerned with creating of software for textural analysis of retinal images by statistic method called co-occurrence matrices as possible alternative method to detection of lesion of retina by the glaucoma diseases. The Glaucoma diseases without well-timed diagnostics and consequential treatment leads to blindness. The retinal images captured by fundus camera are common, easily processable and modern devices are able to make images of 10 Mpix resolution and bigger, what specifies outputs of analytic software. Method of co-occurrence analysis is simple and effective statistic. But computing demandingness, showing up with growing objective quality, comming up as disadvantage.
|
|
|
Classification of arteries and veins in retinal image data
Černohorská, Lucie ; Jakubíček, Roman (referee) ; Kolář, Radim (advisor)
This master's thesis deals with the classification of the retinal blood vessels in retinal image data. The thesis contains a description of anatomy of the human eye with focus on the blood circulation, and imaging and diagnostic methods of the retina are briefly mentioned further. The thesis also summarizes methods of the blood circulation classification with emphasis on the deep learning. The practical section was implemented in Python programming language and describes the pre-processing of the data with determination of AV ratio. Based on a literature search, the U-net architecture was chosen for the classification of the retinal blood vessels. The architecture was modified using the open-source Keras library and tested on images from the experimental video-ophthalmoscope. The modified architecture was initially used for classification of vessels into the corresponding classes and because of unsatisfying results was modified another architecture segmenting retinal vessels, arteries or veins and a proposition of a method of the blood vessels classification.
|
|
|
Bifurcation Localization in Retina Images
Kvapilová, Aneta ; Drahanský, Martin (referee) ; Semerád, Lukáš (advisor)
This thesis deals with processing images of human retina. Its main goal is to create a system which is able to localize places important in a process of creating biometrical template - bifurcations and crossing of blood vessels. The first part focuses on biometrics in detail and explains certain concepts of this area. It also mentions the anatomy of the human eye focusing on retina. The second part provides detailed description of all the stages and algorithms that were necessary in the process of creation of the application.
|
|
|
Statistical analysis of images from optical coherence tomography with retinal applications
Kelča, Jan ; Slávik, Vladimír (referee) ; Kolář, Radim (advisor)
This work briefly describes the anatomy of the eye and retinal structure, including its pathological changes. The work is focused on the characteristics and the principle of optical coherence tomography and its application in ophthalmology in conjunction with statistical analysis of retinal images. The practical part explains problems of statistical analysis retinal images using histograms representing the distribution of gray shades and approximation of functions, which are stretched exponential function and Nakagami function. Suitable optimalization are obtained parameters, from which are formed parametric maps illustrating the distribution of gray levels in different areas of interest.
|
|
|
Detection of the retinal nerve fibre layer
Kunc, Martin ; Szabó, Zoltán (referee) ; Kolář, Radim (advisor)
This thesis is deals with the nerve fibre layer in the colour ophthalmology images of retina. The thesis describes how can we use finding of nerve fibre layer and how was it solved in the past. In the thesis are proposed the methods that are based on processing and scoring frequency spectrums of individual sample of retina. At first here are described the methods of detection on the artificial generated samples that just simulate the nerve fibre layer. Then the thesis concentrates on processing of real images of retina. Because of the bloodstream, that depreciates processing at real images, are all surveyed samples are chosen manually. Except detection the nerve fibre layer itself, the thesis also deals with determination of direction their dissemination.
|
|
|
Manipulator Control for Acquisition of the Eye Retina
Malaník, Petr ; Dvořák, Michal (referee) ; Drahanský, Martin (advisor)
To use the image of the retina for biometric or medical purposes, it is necessary that the image is of the highest quality and ideally covers the largest possible area. It is therefore necessary to move the sensing device to the most suitable position while ensuring its stability. A laboratory handling platform is used for this purpose. The platform is moved by stepper motors. A control board equipped with a microcontroller has been developed for their control, which allows very fine movements. The theoretical accuracy of the manipulator is up to 20 nm. Since it is necessary for sensing, to have proper retina illumination, the control system also includes adjustable illumination elements in the infrared and visible spectrum. The resulting system allows scanning of the retina of the eye from multiple angles and thereby effectively increase the area on which it is possible to further look for diseases or biometric features.
|
|
|
Optic disc detection in video sequences from video ophthalmoscope
Čermák, Marek ; Mézl, Martin (referee) ; Odstrčilík, Jan (advisor)
This work is focused on automatic detection of optic disc in retinal images. There is briefly described anatomy of human eye, principles of retinal imaging and also overview of the methods used for optic disc detection. The practical part describes developed procedures for optic disc detection, ie detection based on watershed transform, active contours and also on region growing technique. The main method of this work is the method of circular transformation, which as the only one allowed to detect the optic disc on the images of video ophtalmoscope and also on the high quality images from fundus cameras. This method was tested on three datasets. The average overlap 92,44 % was achieved for HRF dataset, 91,03 for DRIONS dataset and 77,36 for images of video ophtalmoscope.
|
guest ::
