|
|
Polygonal Mesh Segmentation
Švancár, Matúš ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
This bachelor thesis analyzes and approaches the issue of segmentation of polygonal models. It presents a design of an interactive method inspired by the method described in the Interactive Mesh Segmentation Based on Feature Preserving Harmonic Field. The method uses graph-cut and is implemented as a web application. The application supports .obj and .stl file formats, allows the user to load a model, draw sketches representing foreground and background on the surface of the model, and to start segmentation. Once completed, the user can download the resulting models or continue segmenting with one of them.
|
|
|
Deep Learning for Medical Image Analysis
Szöllösi, Albert ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
This thesis offers possible solution to automatic 3D dental scan landmark localization. These scans are used in dental crown design and digital orthodontics to make the design process easier using specialized software. Before that, though, the scan has to be annotated for the software to know the positions of the teeth. The annotation process is done manually, which guarantees precision, but takes a lot of time. The result of this work could make said process much simpler by applying deep learning. Landmark localization was implemented using a convolutional neural network.
|
|
|
User Interface of FIT Information System
Vyroubal, Marek ; Kodym, Oldřich (referee) ; Beran, Vítězslav (advisor)
The aim of this bachelor thesis is to design and create a user interface (UI) of information system of Faculty of Information Technology, Brno University of Technology which is based on analysis and user testing of current UI. The theoretical part describes UI creation, user friendly design, basic methods of user experience design and about user testing UI. Bachelor thesis focuses on implementation of a web UI or web frontend. The output of this thesis is user research, follow-up to design and implementation of new information system FIT interface based on user testing of the current information system.
|
|
|
Segmentation of hippocampus in MRI data
Kodym, Oldřich ; Chmelík, Jiří (referee) ; Walek, Petr (advisor)
The thesis deals with application of graph-based methods in segmentation of low contrast image data, specifically hippocampus segmentation from magnetic resonance data. Firstly, basics and terminology of graph theory is introduced. Next, minimum graph cut method is explained along with algorithms capable of finding this cut. After that comes the description of its implementation for 2D and 3D image data segmentation. Method was tested on sample data and then implemented as a 3D Slicer software module. Here the method was tested on the hipocampus data of healthy patients as well as patients suffering from Alzheimer’s disease. Most common problems occuring during the segmentation were forshadowed as well as possible ways to solve them.
|
|
|
Volumetric Segmentation of Dental CT Data
Berezný, Matej ; Kodym, Oldřich (referee) ; Čadík, Martin (advisor)
The main goal of this work was to use neural networks for volumetric segmentation of dental CBCT data. As a byproducts, both new dataset including sparse and dense annotations and automatic preprocessing pipeline were produced. Additionally, the possibility of applying transfer learning and multi-phase training in order to improve segmentation results was tested. From the various tests that were carried out, conclusion can be drawn that both multi-phase training and transfer learning showed substantial improvement in dice score for both sparse and dense annotations compared to the baseline method.
|
|
|
Deep Learning for Medical Image Analysis
Bíl, Tomáš ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
The goal of this thesis is developing convolutional neural network which is able to classify if x-ray images are suitable for cephalometry analysis. Four networks were created and trained on a dataset for this purpose. Two of them are VGG type, one is based on UNet and one is Resnet. The dataset was generated from ct scan images. VGG network with four blocks has got the best results. Measured accuracy performed on test dataset is 97%.
|
|
|
Deep Learning for Medical Image Analysis
Dronzeková, Michaela ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
The purpose of this thesis is to use convolutional neural networks for X-ray image classification of human body. Four different architectures of neural networks have been created. They were trained and tested on three tasks: classification of front and lateral chest, classification of X-ray images into several different categories and classification of diseases in chest X-ray. ResNet and SEResNet architectures achieved the best results. SEResNet scored 99,49% accuracy in the first task, ResNet achieved 94,97% accuracy in the second task and SEResNet reached 31,53% in the third task with F1 measure as metrics for evaluating results.
|
|
|
Interactive 3D CT Data Segmentation Based on Deep Learning
Trávníčková, Kateřina ; Hradiš, Michal (referee) ; Kodym, Oldřich (advisor)
This thesis deals with CT data segmentation using convolutional neural nets and describes the problem of training with limited training sets. User interaction is suggested as means of improving segmentation quality for the models trained on small training sets and the possibility of using transfer learning is also considered. All of the chosen methods help improve the segmentation quality in comparison with the baseline method, which is the use of automatic data specific segmentation model. The segmentation has improved by tens of percents in Dice score when trained with very small datasets. These methods can be used, for example, to simplify the creation of a new segmentation dataset.
|
|
|
Text Layout Analysis in Historical Documents
Palacková, Bianca ; Hradiš, Michal (referee) ; Kodym, Oldřich (advisor)
The goal of this thesis is to design and implement algorithm for text layout analysis in historical documents. Neural network was used to solve this problem, specifically architecture Faster-RCNN. Dataset of 6 135 images with historical newspaper was used for training and testing. For purpose of the thesis four models of neural networks were trained: model for detection of words, headings, text regions and model for words detection based on position in line. Outputs from these models were processed in order to determine text layout in input image. A modified F-score metric was used for the evaluation. Based on this metric, the algorithm reached an accuracy almost 80 %.
|
|
|
Material Artefact Generation
Rončka, Martin ; Španěl, Michal (referee) ; Kodym, Oldřich (advisor)
Ne vždy je jednoduché získání dostatečně velké a kvalitní datové sady s obrázky zřetelných artefaktů, ať už kvůli nedostatku ze strany zdroje dat nebo složitosti tvorby anotací. To platí například pro radiologii, nebo také strojírenství. Abychom mohli využít moderní uznávané metody strojového učení které se využívají pro klasifikaci, segmentaci a detekci defektů, je potřeba aby byla datová sada dostatečně velká a vyvážená. Pro malé datové sady čelíme problémům jako je přeučení a slabost dat, které způsobují nesprávnou klasifikaci na úkor málo reprezentovaných tříd. Tato práce se zabývá prozkoumáváním využití generativních sítí pro rozšíření a vyvážení datové sady o nové vygenerované obrázky. Za použití sítí typu Conditional Generative Adversarial Networks (CGAN) a heuristického generátoru anotací jsme schopni generovat velké množství nových snímků součástek s defekty. Pro experimenty s generováním byla použita datová sada závitů. Dále byly použity dvě další datové sady keramiky a snímků z MRI (BraTS). Nad těmito dvěma datovými sadami je provedeno zhodnocení vlivu generovaných dat na učení a zhodnocení přínosu pro zlepšení klasifikace a segmentace.
|
guest ::
