prikaz prve stranice dokumenta Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja
Access restricted to authenticated users
master's thesis
Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja
2022. urn:nbn:hr:179:775235
Ćaleta, Petar
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Cite this document

Ćaleta, P. (2022). Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja (Master's thesis). Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:775235

Ćaleta, Petar. "Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:775235

Ćaleta, Petar. "Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:775235

Ćaleta, P. (2022). 'Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 06 November 2024, https://urn.nsk.hr/urn:nbn:hr:179:775235

Ćaleta P. Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2022 [cited 2024 November 06] Available at: https://urn.nsk.hr/urn:nbn:hr:179:775235

P. Ćaleta, "Višeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja", Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:179:775235

Please login to the repository to save this object to your list.
Metadata
TitleVišeklasna detekcija objekata na slikama prirodnog krajolika upotrebom modela dubokog učenja
Title (english)Multiclass detection of objects in natural landscape images using deep learning models
AuthorPetar Ćaleta
MentorDunja Božić-Štulić (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2022-02-25, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Cilj ovog diplomskog rada bio je izrada modela za višeklasnu detekciju objekata na slikama. Slike su mediteranskog krajolika FESB MLID skupa podataka. Ukupno je 12 klasa koje je moguće detektirati. U radu je objašnjen postupak pripreme slika za treniranje, model koji je korišten, postupak modificiranja modela kako bi se mogao koristiti za višeklasnu detekciju, postupak testiranja podataka te analiza rezultata i točnosti modela. Osim toga, u radu je objašnjena teorija umjetne inteligencije, strojnog i dubokog učenja te način rada neuronskih mreža, te programski jezik Python u kojem je pisan model i pomoćni programi, njegove biblioteke korištene u radu.
Abstract (english)
The aim of this Master Thesis was to create deep learning model for multiclass object detection in images. Images are from FESB MLID data set and depict Mediterranean landscape. There are 12 classes that can be detected on the images. The paper explains the process of preparation of images for training, the model used, the procces of modifying the model so that it can be used for multiclass detection, the process of testing data and analysis of results and accuracy of the model. In addition, the paper explains the theory of artificial intelligence, machine and deep learning and the way neural networks work, as the Python programming language and libraries in which the model and auxiliary programs were written.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:775235
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 conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2022-12-05 11:39:31