prikaz prve stranice dokumenta Prepoznavanje konzumnih riba na slikama postupcima dubokog učenja
Access restricted to authenticated users
master's thesis
Prepoznavanje konzumnih riba na slikama postupcima dubokog učenja
Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. urn:nbn:hr:179:608162
Batinić, Martina
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Institutional repository: Repository of the Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture

Cite this document

Batinić, M. (2022). Prepoznavanje konzumnih riba na slikama postupcima 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:179:608162

Batinić, Martina. "Prepoznavanje konzumnih riba na slikama postupcima 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:179:608162

Batinić, Martina. "Prepoznavanje konzumnih riba na slikama postupcima 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:179:608162

Batinić, M. (2022). 'Prepoznavanje konzumnih riba na slikama postupcima dubokog učenja', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 01 December 2024, https://urn.nsk.hr/urn:nbn:hr:179:608162

Batinić M. Prepoznavanje konzumnih riba na slikama postupcima dubokog učenja [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2022 [cited 2024 December 01] Available at: https://urn.nsk.hr/urn:nbn:hr:179:608162

M. Batinić, "Prepoznavanje konzumnih riba na slikama postupcima 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:179:608162

Metadata
TitlePrepoznavanje konzumnih riba na slikama postupcima dubokog učenja
Title (english)Recognition of edible fish in images using deep learning methods
AuthorMartina Batinić
MentorDarko Stipaničev (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2022-09-14, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Electronics
Abstract
U radu su navedeni i ukratko opisani standardni postupci strojnog učenja za prepoznavanje objekata na slikama. Detaljno su objašnjeni područje dubokog učenja, neuronske mreže, prijenosno učenje te metode klasifikacije slika, detekcije objekata i segmentacije slika. Koristeći programski jezik Python te platformu Tensorflow razvijen je sustav za prepoznavanje konzumnih riba na slikama. Odabran je algoritam konvolucijskih neuronskih mreža. Sustav je razvijen na dva načina: koristeći model konvolucijske neuronske mreže za klasifikaciju obučen od nule te koristeći unaprijed obučeni model mreže VGG16. Tako dobiveni rezultati uspoređeni su s rezultatima standardnih postupaka strojnog učenja dobivenih koristeći istu bazu slika riba s tržnice za obuku.
Abstract (english)
Standard machine learning procedures for recognizing objects in images are listed and briefly described in the paper. The field of deep learning, neural networks, transfer learning and methods of image classification, object detection and image segmentation are explained in detail. Using the Python programming language and the Tensorflow platform, a system was developed to recognize edible fish in images. The algorithm of convolutional neural networks was chosen. The system was developed in two ways: using a convolutional neural network model for classification trained from scratch and using a pre-trained VGG16 network model. The results thus obtained were compared with the results of standard machine learning procedures obtained using the same database of market fish images for training.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:608162
Study programmeTitle: Electronics and Computer Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektronike i računalnog inženjerstva (magistar/magistra inženjer/inženjerka elektronike i računalnog inženjerstva)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2023-02-24 12:04:15