Bc. Roman Ďuriš

Bachelor's thesis

Particle detection in electron microscopy images using neural networks

Particle detection in electron microscopy images using neural networks
Abstract:
Táto práca sa zaoberá segmentačnými metódami snímkov, ktoré používajú neurónové siete, v elektrónovej mikroskopii a vytvára praktickú implementáciu pre tieto modely. Je obzvlášť dôležité, aby tieto metódy boli schopné rozoznať dotýkajúce sa častice a častice, ktoré sa prekrývajú aj za podmienok, ktoré nie sú ideálne. Mask R-CNN, YOLACT a U-Net sú metódy, ktoré sú skúmané ako možné prístupy pre daný …more
Abstract:
This thesis explores image segmentation methods using neural networks for electron microscopy and develops pipelines for those models. Particular emphasis is on the ability of these methods to recognize touching and overlapping particles under non-ideal conditions. Mask R-CNN, YOLACT, and U-Net are explored as viable approaches to the given task. The theoretical part details the selected model’s architecture …more
 
 
Language used: English
Date on which the thesis was submitted / produced: 16. 12. 2021

Thesis defence

  • Date of defence: 8. 2. 2022
  • Supervisor: prof. RNDr. Michal Kozubek, Ph.D.
  • Reader: doc. RNDr. Petr Matula, Ph.D.

Citation record

Full text of thesis

Contents of on-line thesis archive
Published in Theses:
  • světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatiky

Masaryk University

Faculty of Informatics

Bachelor programme / field:
Informatics / Artificial Intelligence and Natural Language Processing