Automatsko otkrivanje umora vozača

master's thesis

• Title (english): Automatic driver drowsiness detection
• Author: Veno Bojanovsky
• Mentor: Igor Sunday Pandžić (mentor)
• Committee member: Igor Sunday Pandžić (predsjednik povjerenstva)
• Committee member: Mario Kušek (član povjerenstva)
• Committee member: Lea Skorin-Kapov (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad iznosi opasnost umora vozača u prometu, koji u moderno vrijeme postaje rješiv pravovremenom detekcijom umora. Kao metoda s najvećim potencijalom otkrivanja umora, u radu se koristi tehnologija praćenja lica, konkretno Visage FaceTracker.
Opisana je problematika vezana uz objektivno ocjenjivanje umora i različite načine na koji se u različitim uvjetima manifestira. Korištenjem parametara nagiba glave, otvorenosti očiju i ustiju, kroz obradu video datoteka s primjerima elemenata umora, pronalaze se optimalne vrijednosti parametara za otkrivanje umora.
Konačno, korištenjem dobivenih parametara se obrađuju video datoteke sa stvarnim primjerima umora te su dani i opisani dobiveni rezultati koji pokazuju razumnu preciznost, obzirom na opseg i kompleksnost problema.

Abstract (english)

The perils of driver fatigue are becoming solvable in modern times, using fatigue detection systems. Given its potential, face tracking technology, the Visage FaceTracker, was explored and used in constructing a driver fatigue detection system.
Problems regarding objectively determining the level of fatigue and the various ways it manifests itself are described. Using head rotation, along with level of eye and mouth openness, as parameters, numerous video files containing elements of fatigue were processed in order to obtain optimal threshold values for the parameters.
Finally, using the optimal values for the parameters, video files containing footage of drowsy and awake people was used to fine tune the values and assert the successfulness of the system. The results show that driver fatigue can be detected using face tracking software, however the degree of its precision warrants more testing fine-tuning.

• Keywords: umor; anotacija; Visage; detekcija; praćenje; lice; vožnja; sigurnost
• Keywords (english): fatigue; drowsiness; annotation; Visage; detection; tracking; face; driving safety
• Language: croatian
• URN:NBN: urn:nbn:hr:168:216484
• Study programme: Title: Information and Communication Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 07:57:24