prikaz prve stranice dokumenta Usporedba različitih modela dubokih neuronskih mreža za detekciju rana
  Download
PDF 2.43 MB
undergraduate thesis
Usporedba različitih modela dubokih neuronskih mreža za detekciju rana
2023. urn:nbn:hr:200:383409
Grčić, Nives
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Computer Engineering and Automation
Chair of Automation and Robotics

Cite this document

Grčić, N. (2023). Usporedba različitih modela dubokih neuronskih mreža za detekciju rana (Undergraduate thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:383409

Grčić, Nives. "Usporedba različitih modela dubokih neuronskih mreža za detekciju rana." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:383409

Grčić, Nives. "Usporedba različitih modela dubokih neuronskih mreža za detekciju rana." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:383409

Grčić, N. (2023). 'Usporedba različitih modela dubokih neuronskih mreža za detekciju rana', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 19 November 2024, https://urn.nsk.hr/urn:nbn:hr:200:383409

Grčić N. Usporedba različitih modela dubokih neuronskih mreža za detekciju rana [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2023 [cited 2024 November 19] Available at: https://urn.nsk.hr/urn:nbn:hr:200:383409

N. Grčić, "Usporedba različitih modela dubokih neuronskih mreža za detekciju rana", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:200:383409

Please login to the repository to save this object to your list.
Metadata
TitleUsporedba različitih modela dubokih neuronskih mreža za detekciju rana
Title (english)Comparison of deep neural network models for wound detection
AuthorNives Grčić
MentorEmmanuel Karlo Nyarko (mentor)
Committee memberGoran Martinović (predsjednik povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Computer Engineering and Automation)
(Chair of Automation and Robotics)
Osijek
Defense date and country2023-09-08, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
Završni rad prikazuje proces primjene modela dubokog učenja na nove podatke. Obrađuju se osnovni principi dubokog učenja te njihova primjena na duboke neuronske mreže. Nadalje, detaljno je opisana arhitektura konvolucijskih neuronskih mreža te pobliže opisana arhitektura dvije konvolucijske neuronske mreže kojima se rad bavi: U-Net i Attention U-Net. Usporedeni su navedeni modeli te evaluirani uporabom DICE koeficijenta. U završnom su radu navedeni prednosti, nedostatci te rezultati usporedbe navedenih modela na augmentiranom skupu podataka na kojem mreže nisu izvorno razvijane.
Abstract (english)
The final thesis presents the process of applying deep learning models to new data. It covers the basic principles of deep learning and their application to deep neural networks. Furthermore, it provides a detailed description of the architecture of convolutional neural networks and specifically describes the architecture of two convolutional neural networks addressed in the thesis: U-Net and Attention U-Net. These models are compared and evaluated using the DICE coefficient. The thesis discusses the advantages, disadvantages, and results of comparing these models on an augmented dataset on which the networks were not originally developed.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:383409
Study programmeTitle: Undergraduate University Study Programme in Computer Engineering
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 resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-10-20 08:21:54