prikaz prve stranice dokumenta Analiza stomatoloških rendgenskih slika pomoću dubokog učenja
No public access
master's thesis
Analiza stomatoloških rendgenskih slika pomoću dubokog učenja
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. urn:nbn:hr:168:440739
Engelmann, Otto
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Engelmann, O. (2021). Analiza stomatoloških rendgenskih slika pomoću dubokog učenja (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:440739

Engelmann, Otto. "Analiza stomatoloških rendgenskih slika pomoću dubokog učenja." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:440739

Engelmann, Otto. "Analiza stomatoloških rendgenskih slika pomoću dubokog učenja." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:440739

Engelmann, O. (2021). 'Analiza stomatoloških rendgenskih slika pomoću dubokog učenja', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 08 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:440739

Engelmann O. Analiza stomatoloških rendgenskih slika pomoću dubokog učenja [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 November 08] Available at: https://urn.nsk.hr/urn:nbn:hr:168:440739

O. Engelmann, "Analiza stomatoloških rendgenskih slika pomoću dubokog učenja", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:440739

Metadata
TitleAnaliza stomatoloških rendgenskih slika pomoću dubokog učenja
Title (english)Analysis of Dental X-ray Images Using Deep Learning Methods
AuthorOtto Engelmann
MentorMarko Subašić (mentor)
Committee memberMarko Subašić (predsjednik povjerenstva)
Committee memberTomislav Petković (član povjerenstva)
Committee memberKrešimir Križanović (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2021-07-01, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Predikcija dobi i spola iz pojedinačnih zuba, paralelno s njihovom detekcijom i klasifikacijom. Korištene su metode dubokog učenja, Faster R-CNN arhitektura i Detectron2 biblioteka. Prvi korak je bila samo detekcija i klasifikacija zuba koja je ostvarena pomoću pristupa sa 16 kategorija. Nakon toga modelu dodani posebni, dodatni slojevi zaduženi za predikciju dobi i spola. Ispostavilo se da treniranje dodatnih slojeva negativno utječe na treniranje slojeva za detekciju te su modeli s njima imali lošije rezultate prilikom detekcije. Konačan model koji je uspješan u svim zadacima istreniran korištenjem par dodatnih tehnika.
Abstract (english)
Prediction of age and gender from individual teeth, in parallel with their detection and classification. In solving this problem there were used deep learning methods, Faster R-CNN architecture and Detectron2 library. The first step was only to detect and classify teeth from images and that was achieved using a 16-category approach. After that, special, additional layers were added to the model with the task of age and gender prediction. It turned out that training additional layers negatively affects training of detection layers and because of that, models with them had worse detection results. Finally, a model that is successful in all tasks was achieved using a couple of additional techniques.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:440739
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 12:11:39