|
|
Volume Rendering Using Programmable HW
Jošth, Radovan ; Kršek, Přemysl (referee) ; Herout, Adam (advisor)
This work describes and implementing method for volume data rendering. Main purpose of this work is visualization of scanned 3D data with some current method used for the 3D volumetric scanning. The 3D volumetric scanning is mainly used in medicine and chemistry. System is using programmable pipeline of current graphic cards, which provides us fast parallel work with large volumetric data. This paper introduces some basics about the volumetric rendering and scanning, describes design and at the end, the implementation steps. Result of this project is application which renders volumetric data with OpenGL.
|
|
|
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.
|
|
|
Rendering Volumetric Data in Web Browser
Fisla, Jakub ; Zemčík, Pavel (referee) ; Španěl, Michal (advisor)
This thesis discusses rendering capabilities of web browsers of accelerated 3D scene rendering. It specifically deals with direct volumetric medical data visualization. It focuses on the usage of ray casting algorithm, its quality and its realistic rendering options. One of the goals was to create an application that demonstrates the ability to render three-dimensional volume data in a web browser using WebGL. The application is written in JavaSript and its 3D rendering core uses the Three.js library.
|
|
|
CPU Rendering of Large Volumetric Data
Majer, Michal ; Milet, Tomáš (referee) ; Španěl, Michal (advisor)
This work examines direct rendering of large volumetric data on the CPU. The aim was to design a parallel implementation of Ray casting algorithm in the Rust programming language and to implement Early Ray Termination and Empty Space Skipping optimizations to speed up rendering. I also designed a demo application to interactively display volumes using this algorithm. A volumetric data generator was also created as part of the work.Both optimizations in the resulting solution offer a 12× speed up. Parallelization further improves this number and renders a large volume at 3.92 FPS on the tested system.
|
|
|
CPU Rendering of Large Volumetric Data
Majer, Michal ; Milet, Tomáš (referee) ; Španěl, Michal (advisor)
This work examines direct rendering of large volumetric data on the CPU. The aim was to design a parallel implementation of Ray casting algorithm in the Rust programming language and to implement Early Ray Termination and Empty Space Skipping optimizations to speed up rendering. I also designed a demo application to interactively display volumes using this algorithm. A volumetric data generator was also created as part of the work.Both optimizations in the resulting solution offer a 12× speed up. Parallelization further improves this number and renders a large volume at 3.92 FPS on the tested system.
|
|
|
Volume data editor
Kavalík, Jiří ; Krajíček, Václav (advisor) ; Kolomazník, Jan (referee)
In the present work we want to design and implement simple tool for basic editing of volume data like resizing, rotating, tresholding retouching, which could be extended as needed. Currently, there are tools for volume data visualisation and simple editors for modding computer games with voxel-based graphics. There is not an editor with basic set of tools for editing volumes in resolution of medical scanners available to common user. The designed and implemented editor can run on common hardware and oth ers set of simple brushes and filters to the user and means to implement diferent tools as plugins for C# programmer.
|
|
|
Vessel segmentation
Dupej, Ján ; Pelikán, Josef (advisor) ; Kolomazník, Jan (referee)
Title: Vessel segmentation Author: Ján Dupej Department / Institute: Department of Software and Computer Science Education Supervisor of the master thesis: RNDr. Josef Pelikán, KSVI Abstract: In this thesis we researched some of the blood vessed segmentation and visualization techniques currently available for angiography on CT data. We then designed, implemented and tested a system that allows both semi-automatic and automatic vessel segmentation and visualization. For vessel segmantation and tracking we used a region-growing algorithm that we overhauled with several heuristics and combined with centerline detection. We then automated this algorithm by automatic seed generation. The visualization part is accomplished with an adaptation of the well-known straightened CPR method that we enhanced so that it visualizes the whole cross-section of the blood vessel, instead of just one line of it. Furthermore, we used the Bishop frame to maintain minimal twist of the curve-local coordinate system along the whole vessel. Keywords: vessel segmentation, medical data analysis, volume data
|
|
|
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.
|
|
|
Hybrid renderer
Goliaš, Matúš ; Pelikán, Josef (advisor) ; Horáček, Jan (referee)
Title: Hybrid renderer Author: Matúš Goliaš Department: Department of Software and Computer Science Education Supervisor: RNDr. Josef Pelikán, KSVI Abstract: In medical applications, there are often used both volume data and objects defined by their surface. However, rendering these two types of data together is rather dif- ficult. This thesis provides the description of ray marching algorithm for volume rendering. We base the algorithm on the emission-absorption model of light trans- fer through a gaseous material. Moreover, we discuss the problems of adding meshes to the volume with respect to the algorithm. Next, we propose an ex- tension of ray marching which takes all meshes into account and renders them with the volume correctly. In addition, we compare the individual approaches for solving steps of the algorithm. Finally, we implement our solution in an ap- plication for volume viewing. We develop this renderer using OpenGL and we discuss our options of implementation. Furthermore, we extend our application by allowing the camera inside the volume, clipping the scene and interactive edit- ing of transfer functions which define the colour representation of volume data. Subsequently, we describe every feature and discuss its possible implementations. Keywords: volume data, ray marching, translucent...
|
|
|
Vessel segmentation
Dupej, Ján ; Pelikán, Josef (advisor) ; Kolomazník, Jan (referee)
Title: Vessel segmentation Author: Ján Dupej Department / Institute: Department of Software and Computer Science Education Supervisor of the master thesis: RNDr. Josef Pelikán, KSVI Abstract: In this thesis we researched some of the blood vessed segmentation and visualization techniques currently available for angiography on CT data. We then designed, implemented and tested a system that allows both semi-automatic and automatic vessel segmentation and visualization. For vessel segmantation and tracking we used a region-growing algorithm that we overhauled with several heuristics and combined with centerline detection. We then automated this algorithm by automatic seed generation. The visualization part is accomplished with an adaptation of the well-known straightened CPR method that we enhanced so that it visualizes the whole cross-section of the blood vessel, instead of just one line of it. Furthermore, we used the Bishop frame to maintain minimal twist of the curve-local coordinate system along the whole vessel. Keywords: vessel segmentation, medical data analysis, volume data
|
guest ::
