Detekcija i raspoznavanje lica u slikama

undergraduate thesis

• Title (english): Face detection and recognition in images
• Author: Tomislav Matić
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Hrvoje Nuić (član povjerenstva)
• Committee member: Zorana Doždor (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Sustavi za detekciju i raspoznavanje lica primjenjivi su u različitim područjima. Razvoj takvih sustava spada u područje računalnog vida. Algoritmi računalnog vida se uvelike oslanjaju na umjetnu inteligenciju odnosno algoritme strojnog učenja, stoga je dan kratak uvod u umjetnu inteligenciju te su obrađeni ključni algoritmi i metode, na kojima se temelji sustav opisan u radu.
Dakle, bilo je potrebno razviti sustav za detekciju i raspoznavanje lica. Sustav je izrađen u programskom jeziku Python i koristi biblioteku face_recognition. Biblioteka omogućava uz pomoć jednostavnog sučelja detektirati i raspoznavati lica s visokom točnošću, a da pri tom ne zahtijeva previše znanja o algoritmima strojnog učenja. Biblioteka u pozadini koristi razne algoritme strojnog učenja, a oni ključni su opisani u radu. Bitan detalj je da biblioteka koristi gotove natrenirane modele visoke točnosti, pa nema potrebe za ponovnim treniranjem.
Sustav radi u stvarnom vremenu tako da se upali web-kamera te se za svaku sličicu snimke provodi postupak detekcije i raspoznavanja lica.

Abstract (english)

Face detection and recognition systems are applicable in various fields. The development of such systems belongs to the field of computer vision. Computer vision algorithms rely heavily on artificial intelligence, i.e. machine learning algorithms, so this is a brief introduction to artificial intelligence and they are processed with key algorithms and methods, on which the system described in the paper is based.
Therefore, it was necessary to develop a system for face detection and recognition. The system was created in the Python programming language and uses the face_recognition library. The library makes it possible to detect and recognize faces with high accuracy with the help of a simple interface, without requiring too much knowledge of machine learning algorithms. The library uses various machine learning algorithms in the background, and the key ones are described in the paper. An important detail is that the library uses ready-made trained models of high accuracy, so there is no need for retraining.
The system works in real time so that the web camera is turned on and the process of face detection and recognition is carried out for each frame of the recording.

• Keywords: detekcija lica; raspoznavanje lica; strojno učenje; duboko učenje; umjetne neuronske mreže
• Keywords (english): face detection; face recognition; machine learning; deep learning; artificial neural networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:671383
• 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: 2023-10-11 09:45:21