Varnostni sistem za nadzor oseb s pomočjo prepoznave obraza

CENTA, JAKA (author), Zrnec, Aljaž (mentor)

V sklopu diplomske naloge smo razvili rešitev za varnostni nadzor nad dogajanjem v opazovanih prostorih. Rešitev izboljša nadzor nad osebami v prostoru ali vhodu v prostor z omejenim dostopom s pomočjo naprednih tehnologij, kot so kognitivne storitve oblačne platforme Microsoft Azure. Rešitev obsega aplikacijo, ki v realnem času zajema slikovne okvirje in jih obdeluje s pomočjo Microsoftovega API-ja za prepoznavo obraza. Obsega tudi spletno aplikacijo za upravljanje sistema in nadzorovanje ter mobilno aplikacijo namenjeno prikazu obvestil o avtoriziranih, neavtoriziranih in neznanih dostopih. Podatke o osebah, lokacijah in dostopih hranimo v podatkovni bazi.

• Language: Slovenian
• Keywords: OpenCV, prepoznava obraza, Face API, Azure, varnostni sistem
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2020
• PID: 20.500.12556/RUL-114743
• COBISS.SI-ID: 1538548931
• Publication date in RUL: 06.03.2020

Secondary language

• Language: English
• Title: Security system for controlling people with the help of face recognition

Abstract

As a part of this thesis, we have developed a solution for controlling events in the observed rooms or spaces. The solution enhances the control of people in the space or entrance to the space with limited access through advanced technologies such as Microsoft Cognitive Services. The solution includes an application that captures image frames in real-time and processes them using the Microsoft Face Recognition API. It also includes an online system management and monitoring application and a mobile application designed to display notifications about authorized, unauthorized and unrecognized accesses. Information about persons, locations and accesses are stored in the database.

• Keywords: OpenCV, Face Recognition, Face API, Azure, Security system