Detekcija kretanja u sustavima video-nadzora

master's thesis

• Title (english): Motion detection in video surveillance systems
• Author: Anita Vencl
• Mentor: Stjepan Bogdan (mentor)
• Committee member: Stjepan Bogdan (predsjednik povjerenstva)
• Committee member: Tamara Petrović (član povjerenstva)
• Committee member: Goran Vasiljević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-02-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

U ovom radu napravljen je program koji može raditi kao programska potpora profesionalnim video nadzornim sustavima ili samostalno uz korištenje kamere bilo koje vrste. Za izradu grafičkog sučelja i svih funkcionalnih dijelova korišteni su Python, OpenCV i Tkinter te razni njima pripadni moduli. Program se sastoji od nekoliko dijelova koje sačinjavaju funkcije koje obrađuju i analiziraju sliku u svrhu tehničke zaštite. Detekcija pokreta po zonama otkriva kretnje u zaštićenim zonama koje određuje korisnik koje su poligonalno proizvoljnih oblika. Detekcija prelaska linije koja detektira objekt koji prelazi virtualnu liniju u okolišu te detekcija ljudi koja pronalazi i obavještava o otkriću ljudskih figura u slici. Program je testiran s više video snimaka i video izvora s kamera u različitim okruženjima.

Abstract (english)

This paper describes creation of a program that can work as supporting software for professional video surveillance systems or independently with the use of any kind of cameras. Python, OpenCV and Tkinter and their various modules were used to create the graphical interface and all functional parts of algorithms. The program consists of several parts made up of functions that process and analyse the image for the purpose of security. Motion detection in zones detects motion in user-defined zones that are polygonal arbitrary in shape. Line crossing detection,which detects an object crossing a virtual line in the environment and people detection,which finds and informs about the discovery of human figures in an image. The program has been tested with multiple videos and video sources from cameras in different environments.

• Keywords: tehnička zaštita; video nadzor; detekcija pokreta; računalni vid; detekcija ljudi; grafičko korisničko sučelje; OpenCV
• Keywords (english): security; video surveillance; motion detection; computer vision; people detection; graphical user interface; OpenCV
• Language: croatian
• URN:NBN: urn:nbn:hr:168:277095
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 14:08:10