prikaz prve stranice dokumenta Automatsko vizualno očitavanje brojila pomoću dubokog učenja
No public access
master's thesis
Automatsko vizualno očitavanje brojila pomoću dubokog učenja
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. urn:nbn:hr:168:900561
Koščević, Karlo
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Koščević, K. (2018). Automatsko vizualno očitavanje brojila 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:168:900561

Koščević, Karlo. "Automatsko vizualno očitavanje brojila pomoću dubokog učenja." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. https://urn.nsk.hr/urn:nbn:hr:168:900561

Koščević, Karlo. "Automatsko vizualno očitavanje brojila pomoću dubokog učenja." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. https://urn.nsk.hr/urn:nbn:hr:168:900561

Koščević, K. (2018). 'Automatsko vizualno očitavanje brojila pomoću dubokog učenja', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 03 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:900561

Koščević K. Automatsko vizualno očitavanje brojila pomoću dubokog učenja [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2018 [cited 2024 June 03] Available at: https://urn.nsk.hr/urn:nbn:hr:168:900561

K. Koščević, "Automatsko vizualno očitavanje brojila pomoću dubokog učenja", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2018. Available at: https://urn.nsk.hr/urn:nbn:hr:168:900561

Metadata
TitleAutomatsko vizualno očitavanje brojila pomoću dubokog učenja
Title (english)Automatic Visual Reading of Meters Using Deep Learning
AuthorKarlo Koščević
MentorMarko Subašić (mentor)
Committee memberMarko Subašić (predsjednik povjerenstva)
Committee memberSven Lončarić (član povjerenstva)
Committee memberTomislav Petković (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2018-07-13, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovom radu dan je pristup očitanja komunalnih brojila uporabom algoritma dubokog učenja. Pri tome je kao početna točka korišten algoritam Faster R-CNN te je njegova funkcionalnost modificirana kako bi se ostvarilo što preciznije očitanje. Za rješavanje ovog problema bilo je potrebno prikupiti bazu slika i organizirati označeni skup podataka za učenje, a zbog povećanja skupa podataka provedena je i dodatna augmentacija. Tako je u ovom radu dan i opis augmentacijskih tehnika, ali i modifikacija za automatsko augmentiranje označenog skupa podataka. Za svaki dio metode za očitanje komunalnih brojila pa tako i cijelu metodu provedeni su eksperimenti koji pokazuju uspješnost razvijenog sustava.
Abstract (english)
These thesis presents an novel approach to the problem of reading of residential meters using deep learning algorithms. As a starting point Faster R-CNN method was used and, to acquire more precise readings, it's functionality was modified. As there were no databases for this kind of task, one had to be collected and properly annotated. This thesis also provide a brief introduction to methods for image augmentation and a technique to augment annotated image dataset. For each module of presented method as well as the whole method as one unit some experiments were conducted to show the successfulness of the system.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:900561
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 on2019-03-13 21:46:02