Thesis Details

Zvýšení rozlišení obrazu pomocí hlubokých neuronových sítí

Master's Thesis Student: Mojžiš Tomáš Academic Year: 2021/2022 Supervisor: Španěl Michal, Ing., Ph.D.
Language
Slovak
Abstract

The aim of this thesis is to create a deep neural net capable of super-resolution on images acquired by electron microscopes. The thesis consists of two parts - finding appropriate data and creating a dataset for the super-resolution task and designing a neural net architecture capable of solving the super-resolution task. Within the thesis, two datasets comprised of images acquired by electron microscopes were created. The datasets differ in the approach to data augmentation. They allow to train a neural network which fulfills the super-resolution task. To solve this task, two U-Net based and one GAN based architecture were trained. The resolution of images was upscaled by a factor of two and four. The best artificially upscaled images were created by neural network Real-ESRGAN. The values of metrics were not higher than the tested interpolation method, but the images seem more visually pleasing especially when they were upscaled four times. Thanks to this thesis, two datasets were created allowing to train other possible neural network architectures to improve the quality of the artificially upscaled images. The neural networks trained in this thesis can be utilized in the process of acquiring higher quality data from low resolution electron microscope images.

Keywords

deep neural network, convolutional neural network, deep learning, computer vision, super-resolution, electron microscopy, gan, unet, dataset

Department
Degree Programme
Information Technology and Artificial Intelligence, Specialization Intelligent Systems
Files
Status
defended, grade A
Date
20 June 2022
Reviewer
Committee
Zbořil František V., doc. Ing., CSc. (DITS FIT BUT), předseda
Bidlo Michal, doc. Ing., Ph.D. (DCSY FIT BUT), člen
Janoušek Vladimír, doc. Ing., Ph.D. (DITS FIT BUT), člen
Kanich Ondřej, Ing., Ph.D. (DITS FIT BUT), člen
Peringer Petr, Dr. Ing. (DITS FIT BUT), člen
Smrž Pavel, doc. RNDr., Ph.D. (DCGM FIT BUT), člen
Citation
MOJŽIŠ, Tomáš. Zvýšení rozlišení obrazu pomocí hlubokých neuronových sítí. Brno, 2022. Master's Thesis. Brno University of Technology, Faculty of Information Technology. 2022-06-20. Supervised by Španěl Michal. Available from: https://www.fit.vut.cz/study/thesis/24493/
BibTeX
@mastersthesis{FITMT24493,
    author = "Tom\'{a}\v{s} Moj\v{z}i\v{s}",
    type = "Master's thesis",
    title = "Zv\'{y}\v{s}en\'{i} rozli\v{s}en\'{i} obrazu pomoc\'{i} hlubok\'{y}ch neuronov\'{y}ch s\'{i}t\'{i}",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2022,
    location = "Brno, CZ",
    language = "slovak",
    url = "https://www.fit.vut.cz/study/thesis/24493/"
}
Back to top