prikaz prve stranice dokumenta PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU
  Download
PDF 774.17 KB
undergraduate thesis
PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU
2019. urn:nbn:hr:200:321699
Budak, Luka
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Communications
Chair of Multimedia Systems and Digital Television

Cite this document

Budak, L. (2019). PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU (Undergraduate thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:321699

Budak, Luka. "PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2019. https://urn.nsk.hr/urn:nbn:hr:200:321699

Budak, Luka. "PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2019. https://urn.nsk.hr/urn:nbn:hr:200:321699

Budak, L. (2019). 'PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 28 December 2024, https://urn.nsk.hr/urn:nbn:hr:200:321699

Budak L. PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2019 [cited 2024 December 28] Available at: https://urn.nsk.hr/urn:nbn:hr:200:321699

L. Budak, "PREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:200:321699

Please login to the repository to save this object to your list.
Metadata
TitlePREPOZNAVANJE OSOBA U SLICI ZASNOVANO NA DUBOKOM UČENJU
Title (english)RECOGNIZING PERSONS IN AN IMAGE BASED ON DEEP LEARNING
AuthorLuka Budak
MentorMario Vranješ (mentor)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Communications)
(Chair of Multimedia Systems and Digital Television)
Osijek
Defense date and country2019-09-26, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
U ovom radu objašnjeni su principi rada dubokog učenja za rješavanje problema prepoznavanja osobe u digitalnoj slici. Općenito se za rješavanje problema u računalnom vidu koriste konvolucijske neuronske mreže. Ovaj rad se bavi prilagođavanjem postojeće konvolucijske neuronske mreže koja će obavljati prepoznavanje osoba. Tehnikom prenesenog učenja na VGGNet-16 modelu je izgrađena mreža koja može prepoznati osam jedinstvenih identiteta. Treniranje je obavljeno na javno dostupnoj bazi podataka slika lica čiji se broj slika dodatno povećao augmentacijom skupa tih slika. Dio slika iz baze podataka se izdvojio od trening podataka za kasnije vrednovanje dobivenog modela. Vrednovanjem modela na tim slikama dobije se prosječna točnost prepoznavanja od 83 posto.
Abstract (english)
This paper explains the principles of deep learning work to solve the problem of recognizing a person in a digital image. Generally, convolutional neural networks are used to solve computer vision problems. This paper deals with the adjustment of an existing convolutional neural network that will perform recognition of persons. Using transfer learning on VGGNet-16 model a network has been built that can recognize eight unique identities. The training was performed on a publicly available database of facial images whose number of images was further increased by augmenting a set of these images. Part of the images from the database have been sorted out from the training data for later evaluation of the made model. Evaluating the models with these images yields an average recognition accuracy of 83 precent.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:321699
Study programmeTitle: Undergraduate University Study Programme
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2019-10-28 13:45:01