prikaz prve stranice dokumenta Analiza ultrazvučnih slika autoenkoderskim mrežama
No public access
master's thesis
Analiza ultrazvučnih slika autoenkoderskim mrežama
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. urn:nbn:hr:168:711526
Milković, Fran
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Milković, F. (2020). Analiza ultrazvučnih slika autoenkoderskim mrežama (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:711526

Milković, Fran. "Analiza ultrazvučnih slika autoenkoderskim mrežama." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:711526

Milković, Fran. "Analiza ultrazvučnih slika autoenkoderskim mrežama." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:711526

Milković, F. (2020). 'Analiza ultrazvučnih slika autoenkoderskim mrežama', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 27 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:711526

Milković F. Analiza ultrazvučnih slika autoenkoderskim mrežama [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2020 [cited 2024 December 27] Available at: https://urn.nsk.hr/urn:nbn:hr:168:711526

F. Milković, "Analiza ultrazvučnih slika autoenkoderskim mrežama", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:168:711526

Metadata
TitleAnaliza ultrazvučnih slika autoenkoderskim mrežama
Title (english)Ultrasound Image Analysis Using Autoencoder Networks
AuthorFran Milković
MentorSven Lončarić (mentor)
Committee memberSven Lončarić (predsjednik povjerenstva)
Committee memberMarko Subašić (član povjerenstva)
Committee memberTomislav Petković (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2020-07-10, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Ultrazvučno snimanje jedna je od najkvalitetnijih metoda nedestruktivnog testiranja koja pruža uvid u unutarnje strukture i procese snimanih objekata. Uobičajena praksa je da stručnjaci pregledavaju ultrazvučne snimke te donose zaključke o stanju promatranog objekta. U ovom radu istražena je primjena varijacijskog autoenkodera kao dijela sustava za asistiranu analizu i detekciju anomalija na ultrazvučnim snimkama blokova tipičnih za nuklearna postrojenja. Postignuti su rezultati poput pronalaska iznad 96% anomalija uz stopu lažnih detekcija od samo 13%. Istražena je i primjena različitih dimenzija ulaznih slika kao i RGB slika te integracija dubokih i predtreniranih modela u strukturu varijacijskog autoenkodera.
Abstract (english)
Ultrasound imaging is one of the highest quality methods of non-destructive testing which gives insight into inner structures and processes of observed objects. Standard procedure is having experts inspect the obtained ultrasound images and arrive to conclusions about the object's condition. This thesis explores the application of variational autoencoders as a part of a system for assisted analysis and anomaly detection in ultrasound images of blocks typical for nuclear facilities. Achieved results are finding over 96% of all anomalies with a false positive rate of as low as 13%. The use of different input image sizes, RGB images, deep and pretrained models was also explored.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:711526
Study programmeTitle: Information and Communication Technology
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 13:13:28