prikaz prve stranice dokumenta Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS
No public access
master's thesis
Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. urn:nbn:hr:168:994986
Bukić, Ivan
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Bukić, I. (2024). Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:994986

Bukić, Ivan. "Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:994986

Bukić, Ivan. "Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:994986

Bukić, I. (2024). 'Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 17 February 2025, https://urn.nsk.hr/urn:nbn:hr:168:994986

Bukić I. Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2024 [cited 2025 February 17] Available at: https://urn.nsk.hr/urn:nbn:hr:168:994986

I. Bukić, "Prepoznavanje emocija u stvarnom vremenu na robotu TIOSS", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:168:994986

Metadata
TitlePrepoznavanje emocija u stvarnom vremenu na robotu TIOSS
Title (english)Real-time emotion recognition on the TIOSS robot
AuthorIvan Bukić
MentorAna Sović Kržić (mentor)
Committee memberAna Sović Kržić (predsjednik povjerenstva)
Committee memberJuraj Petrović (član povjerenstva)
Committee memberDamir Seršić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2024-07-04, Croatia
Abstract
Prepoznavanje emocija na licu je ključno za uspješnu komunikaciju među ljudima i može biti primjenjeno u različitim područjima, npr. interakcija čovjeka i robota ili sigurnosne svrhe. Cilj rada je povezivanje automatskog prepoznavanja emocija s robotom TIOSS-om, prvim hrvatskim humanoidnim robotom, izrađenim 1958. godine. Robot je opremljen kamerom i Raspberry Pi 5 računalom. Kroz rad su istraženi najbolji postojeći algoritmi prepoznavanja emocija kao što su različiti tipovi konvolucijskih neuronskih mreža (CNN). Algoritmi su trenirani na standardnim skupovima podataka, FER-2013, CK+ i RAF-DB, koji sadrže tisuće označenih slika lica s različitim emocionalnim izrazima. Odabrani algoritmi su prilagođeni za korištenje na Raspberry Pi 5 računalu u stvarnom vremenu. Algoritmi su testirani i implementirani na TIOSS-u, omogućujući robotu različite glasovne reakcije na temelju prepoznatih emocija.
Abstract (english)
Facial emotion recognition is crucial for successful human communication and can be applied in various fields such as human-robot interaction or security purposes. The objective of this work is to integrate automatic emotion recognition with TIOSS, the first Croatian humanoid robot, created in 1958. The robot is equipped with a camera and a Raspberry Pi 5 computer. Throughout this work, the best existing emotion recognition algorithms, such as convolutional neural networks (CNN), were explored. These algorithms were trained on standard datasets like FER-2013 and RAF-DB, which contain thousands of labeled facial images with different emotional expressions. The selected algorithms were then adapted for real-time use on the Raspberry Pi 5 computer. The algorithms were tested and implemented on TIOSS, enabling the robot to generate different voice reactions based on the recognized emotions.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:994986
Study programmeTitle: Information and Communication Technology
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2024-10-12 20:14:20