Detekcija emocija djece iz fotografije

undergraduate thesis

• Title (english): Emotion detection of children from photography
• Author: Lucija Medić
• Mentor: Ana Sović Kržić (mentor)
• Committee member: Ana Sović Kržić (predsjednik povjerenstva)
• Committee member: Maja Jurišić Bellotti (član povjerenstva)
• Committee member: Ivan Ralašić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

U ovom radu su implementirane metode za prepoznavanje lica i detektiranje emocija lica na fotografijama koristeći klasifikaciju Gaborovim filtrom i različite algoritme, Viola-Jones algoritam i analizu glavnih komponenti. Navedeni filtar i algoritmi su u teorijskom dijelu objašnjeni, a u praktičnom dijelu su prikazani rezultati istih provedeni na fotografijama djece. Kombinirajući Gaborov filtar, neuronske mreže i backpropagation algoritam detektirana su lica na fotografiji. Emocije su u ovom radu detektirane koristeći dva algoritma. Analizom glavnih komponenti emocije su uspješno i brzo detektirane te je pritom korištena javno dostupna baza slika. Algoritmom Viola-Jones emocije pojedine slike su detektirane bez skupa za učenje, što je veoma efikasno, iako je broj detektiranih emocija ograničen.

Abstract (english)

In this paper, methods for facial recognition and facial expression detection are performed using the Gabor filter classification and various algorithms, the Viola-Jones algorithm and the analysis of the main components. The aforementioned filter and algorithms are explained in the theoretical part, and in the practical part the results of the same are presented in the photographs of the children. By combining Gabor's filter, neural network and backpropagation algorithm, faces were detected in the photo. Emotions in this paper are revealed by two algorithms. By analyzing the main components of the emotion, it was successfully and quickly discovered, and a publicly available image base was used. The Viola-Jones algorithm of the emotions of individual images has been uncovered without a learning set, which is very effective, although the number of detected emotions is limited.

• Keywords: emocije; detekcija lica; detekcija emocija; Matlab; Gaborov filtar; Viola-Jones algoritam; Analiza glavnih komponenti
• Keywords (english): emotions; face detection; Emotion detection; Matlab; Gabor's filter; Viola-Jones algorithm; Major component analysis
• Language: croatian
• URN:NBN: urn:nbn:hr:168:490338
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 22:55:05