Detekcija i raspoznavanje lica u slikama

undergraduate thesis

• Title (english): Face detection and recognition in images
• Author: Josip Srbljić
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Lea Skorin-Kapov (član povjerenstva)
• Committee member: Darijan Marčetić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Detekcija i raspoznavanje lica u slikama tehnologija je koja se istražuje više od 50 godina. Završni rad podijeljen je u dvije velike cjeline, detekcija i raspoznavanje lica. U prvom dijelu rada obrađene su teme značajki, pojačavanja i kaskadnih klasifikatora koje se koriste za detekciju lica. U cjelini posvećenoj raspoznavanju lica istražene su metode reprezentacije lica na temelju svojstvenih lica (eigenfaces), Fisherovih lica (Fisherfaces) i Lokalnih binarnih uzoraka (local binary patterns). Klasifikator je u sva tri slučaja bio temeljen na metodi najbližeg susjeda (1-NN). Ispitivanje metoda provedeno je sustavom za detekciju i identifikaciju osoba u slikama koji sam razvio koristeći biblioteku OpenCV. Sustav na temelju unaprijed pripremljenog skupa slika lica osoba, detektira i identificira osobe u nekom skupu slika, npr. albumu.

Abstract (english)

The thesis is divided in to two parts: face detection and face recognition. The first part of the thesis describes algorithms and techniques needed to preform face detection in pictures. Algorithms Viola Jones and Rainer Lienhart for object detection and cascade training algorithm AdaBoost are described. The face recognition part of the thesis describes methods used in face recognition systems and applications. Algorithms Eigenfaces, Fisherfaces and Local Binary patterns are described and implemented in a system that detects and recognizes faces in pictures. Each face recognition method is combined with a classifier based on a nearest neighbor algorithm (1 - NN) [26]. Face detection and recognition system has been developed using OpenCV libraries. System performance is presented in standard biometric values for each face recognition method in controlled and real time conditions.

• Keywords: Detekcija lica; Raspoznavanje lica; Svojstvena lica; Fisherfaces; Lokalne binarne značajke; Viola Jones
• Keywords (english): Face Detection; Face recognition; Eigenfaces; Fisherfaces; Local binary patterns; Viola Jones
• Language: croatian
• URN:NBN: urn:nbn:hr:168:854699
• 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: 2018-02-06 11:53:47