Nadgledanje pristupa prostoru pomoću raspoznavanja lica

master's thesis

• Title (english): Access Control Based on Face Recognition
• Author: Jakov Raguž
• Mentor: Igor Sunday Pandžić (mentor)
• Committee member: Igor Sunday Pandžić (predsjednik povjerenstva)
• Committee member: Mirko Sužnjević (član povjerenstva)
• Committee member: Lea Skorin-Kapov (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Sustav za raspoznavanje lica u ovom radu implementiran je koristeći programski jezik Python te primarno biblioteku OpenCV. Implementaciju možemo podijeliti na pet koraka: prikupljanje podataka, treniranje modela, validacija modela, testiranje modela i izrada korisničkog sučelja. Za treniranje modela prvo je potrebno prikupiti fotografije osoba koje želimo da naš model nauči prepoznati. Nakon prikupljanja fotografija slijedi treniranje modela LBPH algoritmom dostupni također u OpenCV biblioteci. Sljedeći korak je validacija treniranog modela s ciljem određivanja vrijednosti praga za mjeru udaljenosti. Kada je prag određen slijedi izrada programa koji omogućava raspoznavanje osoba u stvarnom vremenu. Kako bi se korisniku pojednostavila upotreba sustava, na kraju se kreira korisničko sučelje. Raspravljeni su glavni problemi prilikom implementacije kao i moguća poboljšanja.

Abstract (english)

The face recognition system in this paper is implemented using the Python programming language and primarily the OpenCV library. Implementation can be divided into five steps: data collection, model training, model validation, model testing, and user interface development. To train a model, it is first necessary to collect photos of the people we want our model to learn to recognize. The collection of photographs is followed by training of the model with the LBPH algorithm, also available in the OpenCV library. The next step is to validate the trained model in order to determine the threshold value for the distance measure. Once the threshold has been set, a program is created that enables the recognition of people in real-time. In order to make the system user-friendly, a user interface is created. The main problems during implementation as well as possible improvements were also discussed.

• Keywords: Python; OpenCV; Dlib; detekcija lica; raspoznavanje lica; treniranje modela
• Keywords (english): Python; OpenCV; Dlib; face detection; face recognition; model training
• Language: croatian
• URN:NBN: urn:nbn:hr:168:162596
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 09:48:29