prikaz prve stranice dokumenta Detekcija objekata u zračnim snimkama temeljena na dubokom učenju
No public access
master's thesis
Detekcija objekata u zračnim snimkama temeljena na dubokom učenju
2021. urn:nbn:hr:168:622239
Tokić, Marin
University of Zagreb
Faculty of Electrical Engineering and Computing

Cite this document

Tokić, M. (2021). Detekcija objekata u zračnim snimkama temeljena na dubokom učenju (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:622239

Tokić, Marin. "Detekcija objekata u zračnim snimkama temeljena na dubokom učenju." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:622239

Tokić, Marin. "Detekcija objekata u zračnim snimkama temeljena na dubokom učenju." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:622239

Tokić, M. (2021). 'Detekcija objekata u zračnim snimkama temeljena na dubokom učenju', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 06 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:622239

Tokić M. Detekcija objekata u zračnim snimkama temeljena na dubokom učenju [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 December 06] Available at: https://urn.nsk.hr/urn:nbn:hr:168:622239

M. Tokić, "Detekcija objekata u zračnim snimkama temeljena na dubokom učenju", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:622239

Please login to the repository to save this object to your list.
Metadata
TitleDetekcija objekata u zračnim snimkama temeljena na dubokom učenju
Title (english)Object Detection in Aerial Images Based on Deep Learning
AuthorMarin Tokić
MentorZoran Kalafatić (mentor)
Committee memberZoran Kalafatić (predsjednik povjerenstva)
Committee memberTomislav Hrkać (član povjerenstva)
Committee memberVladimir Čeperić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2021-02-24, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Detekcija objekata u satelitskim snimkama je kompleksan problem. Proizvoljna orijentacija objekata te velika varijacija u veličini pojedinih objekata predstavlja problem modelima koji se koriste sidrištima. U radu su analizirani načini rada dvaju dubokih modela koji nastoje popraviti te nedostatke. Testirano je ponašanje modela prilikom treniranja raznim optimizatorima. Za treniranje i evaluaciju modela korišten je skup podataka DOTA.
Abstract (english)
Detection of objects in aerial images is a complex problem. Arbitrary orientation of objects and large variation in sizes of individual objects make detection with anchor based methods difficult. We analyze the modes of operation of two deep neural networks that seek to fix these shortcomings. We test how they behave when we train them with different optimizers. We use the DOTA dataset for testing and evaluation purposes.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:622239
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 14:01:25