Automatsko raspoznavanje osmjeha iz slike lica

undergraduate thesis

• Title (english): Automatic smile recognition from facial image
• Author: Dario Jaić
• Mentor: Igor Sunday Pandžić (mentor)
• Committee member: Igor Sunday Pandžić (predsjednik povjerenstva)
• Committee member: Miroslav Frljak (član povjerenstva)
• Committee member: Nenad Markuš (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2016-09-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Korištena su dva algoritma za raspoznavanje osmjeha iz slike lica.
Jednostavniji algoritam računa ako je udaljenost manja od zadanog parametra, onda se osoba ne smije, a ako je veća osoba se smije. Metoda se temelji na činjenici da, kada se osoba smije, usta se šire i ekstremne točke ustiju se približavaju očima.
Složeni algoritam radi po principu određivanja tzv. Patcheva na licu, dobavljanja odgovarajućih podataka Local Binary Pattern (LBN), računanja histograma (59-bin), te sastavljanja feature vectora za tu sliku. Nakon toga se trenira tzv. Support Vector Machine pomoću kojega računalo "nauči" razlikovati feature vektore slika s osmijehom i slika bez osmijeha.

Abstract (english)

Two algorithms were used for smile recognition from the facial image.
A simpler algorithm account if the distance is less than a given parameter, then the person is not smiling, and if it is higher person is smiling. The method is based on the fact that, when a person laughs, mouth spread and extreme points of mouth are approaching eyes.
A complex algorithm works on the principle of determining the so-called. Patches on the face, they supply the appropriate data Local Binary Pattern (LBN), computing a histogram (59-bin), and making Feature Vector for the image. Then the Support Vector Machine is being trained with which the computer "learn" to distinguish between feature vectors picture with a smile and without smile.

• Keywords: Algoritam vektora potpore; OpenCV; Lokalni binarni uzorak; histogram; Svojstvena točka; svojstven vektor; FDP
• Keywords (english): Support Vector Machine (SVM); OpenCV; Local Binary Pattern(LPB); histogram; Feature Vector; Feature Point; FDP
• Language: croatian
• URN:NBN: urn:nbn:hr:168:487920
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-19 14:28:56