Segmentacija trodimenzionalnih slika srca dubokom neuronskom mrežom

undergraduate thesis

• Title (english): Segmentation of Three-Dimensional Cardiac Images Using Deep Neural Network
• Author: Ardita Šalja
• Mentor: Sven Lončarić (mentor)
• Committee member: Sven Lončarić (predsjednik povjerenstva)
• Committee member: Jelena Ivanković (član povjerenstva)
• Committee member: Vedran Stipetić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U sklopu ovog rada objašnjene su umjetne neuronske mreže s naglaskom na konvolucijske duboke neuronske mreže. Objašnjen je proces učenja neuronske mreže kao i korištene funkcije gubitka, regularizacije i problemi pri učenju. Predstavljeni su hiperparametri i optimalan odabir istih. Odabrana je prigodna arhitektura koja se temelji na konvolucijskim mrežama za problem semantičke segmentacije te je izvedena i modificirana implementacije iste. Na kraju rada prikazani su rezultati i moguće daljnje preinake za kvalitetniju segmentaciju.

Abstract (english)

In this paper we introduced artificial neural networks with focus on deep convolutional neural networks. We described the learning process of a neural network as well as loss functions, regularizations and problems faced in the process of learning. We introduced hyperparameters and choosing their optimal values. We chose suitable architecture based on convolutional neural networks for problem of semantic segmentation and we implemented given architecture. In the end we presented the results and possible modifications for better segmentation.

• Keywords: računalni vid; semantička segmentacija; neuronske mreže; konvolucijska neuronska mreža; biomedicinske slike
• Keywords (english): computer vision; semantic segmentation; neural networks; convolutional neural network; biomedical images
• Language: croatian
• URN:NBN: urn:nbn:hr:168:341793
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 22:15:25