prikaz prve stranice dokumenta Automatsko otkrivanje umora vozača
No public access
master's thesis
Automatsko otkrivanje umora vozača
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. urn:nbn:hr:168:278307
Pezelj, Dominik
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Pezelj, D. (2022). Automatsko otkrivanje umora vozača (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:278307

Pezelj, Dominik. "Automatsko otkrivanje umora vozača." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. https://urn.nsk.hr/urn:nbn:hr:168:278307

Pezelj, Dominik. "Automatsko otkrivanje umora vozača." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. https://urn.nsk.hr/urn:nbn:hr:168:278307

Pezelj, D. (2022). 'Automatsko otkrivanje umora vozača', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 27 September 2024, https://urn.nsk.hr/urn:nbn:hr:168:278307

Pezelj D. Automatsko otkrivanje umora vozača [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2022 [cited 2024 September 27] Available at: https://urn.nsk.hr/urn:nbn:hr:168:278307

D. Pezelj, "Automatsko otkrivanje umora vozača", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:168:278307

Metadata
TitleAutomatsko otkrivanje umora vozača
Title (english)Automatic Driver Drowsiness Detection
AuthorDominik Pezelj
MentorIgor Sunday Pandžić (mentor)
Committee memberIgor Sunday Pandžić (predsjednik povjerenstva)
Committee memberMirko Sužnjević (član povjerenstva)
Committee memberLea Skorin-Kapov (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2022-07-13, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Umor vozača je važan faktor smanjenja sigurnosti u prometu. Pokazuje se da umor smanjuje kognitivne sposobnosti, ponajprije procjenu i percepciju okoline, a istovremeno je jedan od značajnijih uzroka prometnih nesreća sa smrtnim ishodom, posebno prisutan kod profesionalnih vozača. Procjenjuje se da bi broj vozila u prometu do 2040. godine mogao biti dvostruko veći zbog čega sustavi automatskog otkrivanja umora mogu pridonijeti smanjenju broja prometnih nesreća uzrokovanih umorom vozača. Sustavi otkrivanja umora procjenjuju umor temeljem prikupljenih informacija. U okviru rada opisana su i implementirana dva sustava automatskog otkrivanja umora čiji se rad zasniva na mogućnosti detekcije, praćenja i analize ljudskog lica na slici. Za pronalazak lica i izračun specifičnih točaka lica korištene su implementacije biblioteke dlib unutar Python okruženja. Pomoću specifičnih točaka lica izračunavaju se EAR (engl. Eye Aspect Ratio) i MAR (engl. Mouth Aspect Ratio). Ove mjere omogućuju otkrivanje zatvorenosti očiju ili otvorenosti usta vozača na slici. Analizom videozapisa u stvarnom vremenu moguće je utvrditi primjerice ima li vozač zatvorene oči, trepće li, koja je frekvencija treptanja, zijeva li te koja je frekvencija zijevanja. Ove se informacije mogu iskoristiti za procjenu umora vozača. Drugi je pristup direktno korištenje EAR i MAR vrijednosti i modela strojnog učenja. Vrijednosti se proslijede naučenom modelu koji zatim procjenjuje umor vozača.
Abstract (english)
Driver drowsiness detection is a car safety technology that helps prevent accidents caused by the driver getting drowsy. Various studies have suggested that a high amount of all road accidents are fatigue-related. In this paper, two automatic driver drowsiness detection systems are explained based on the possibility of detection, tracking, and analysis of the human face in the image. For face detection, dlib's HOG face detector is used along with the facial landmarks detector. Using the facial landmarks, eye aspect ratio - EAR, and the mouth aspect ratio - MAR can be calculated. EAR and MAR detect if the driver has closed eyes or an open mouth in a picture. Analyzing these measurements on every frame of the video feed makes it possible to detect continuous eye closure, blinks, yawns, etc. Based on that information, driver drowsiness estimation can be made. Another approach is using a machine learning model that directly receives EAR and MAR values for driver drowsiness estimation.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:278307
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 10:35:58