prikaz prve stranice dokumenta Algoritmi za detekciju umora vozača
No public access
master's thesis
Algoritmi za detekciju umora vozača
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2015. urn:nbn:hr:168:079058
Mandić, Zvonimir
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Mandić, Z. (2015). Algoritmi za detekciju umora vozača (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:079058

Mandić, Zvonimir. "Algoritmi za detekciju umora vozača." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2015. https://urn.nsk.hr/urn:nbn:hr:168:079058

Mandić, Zvonimir. "Algoritmi za detekciju umora vozača." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2015. https://urn.nsk.hr/urn:nbn:hr:168:079058

Mandić, Z. (2015). 'Algoritmi za detekciju umora vozača', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 27 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:079058

Mandić Z. Algoritmi za detekciju umora vozača [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2015 [cited 2024 December 27] Available at: https://urn.nsk.hr/urn:nbn:hr:168:079058

Z. Mandić, "Algoritmi za detekciju umora vozača", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2015. Available at: https://urn.nsk.hr/urn:nbn:hr:168:079058

Metadata
TitleAlgoritmi za detekciju umora vozača
Title (english)Algorithms for detecting driver fatigue
AuthorZvonimir Mandić
MentorDavor Petrinović (mentor)
Committee memberDavor Petrinović (predsjednik povjerenstva)
Committee memberDamir Seršić (član povjerenstva)
Committee memberMarko Subašić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2015-07-10, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U radu je opisan sustav za detekciju umora baziran na analizi zijevanja. Sustav uzima koordinate vrha gornje usne i dna brade iz čega računa njihovu udaljenost. Ta udaljenost se zatim provlači kroz sklop za normalizaciju i skaliranje u interval od 0 do 1 nakon čega se iz signala filtrom uklanjaju nepotrebne informacije. Iz filtriranog signala sklopom za detekciju lokalnog maksimuma izdvajaju se vršne vrijednosti. Detektor zijeva zatim uzima te vršne vrijednosti, uspoređuje ih sa zadanim pragom te signalizira ukoliko je zijev detektiran. Brojač vremena pamti trenutke kada je zijevanje detektirano te na izlaz daje intervale između svakog zijeva. Pomoću tih intervala određuje se frekvencija zijevanja. Zadavanjem praga frekvencije zijevanja kod umorne osobe ostvaruje se detekcija umora vozača. Detekcija bazirana na analizom zijevanja je funkcionalna i može se koristiti u stvarnim uvjetima. No, takav sustav predviđen je kao dio većeg sustava koji će donositi odluke bazirane na više različitih izvora informacija. Osim praktične implementacije, u radu su opisani i istraživački radovi na temu detekcije umora vozača. U tim radovima detekcija se bazira na EEG-u, analizi pritiska na volan i analizi karakteristika lica čemu je dana posebna pažnja. U tim radovima pokazano je da je EEG najtočnija metoda za detekciju umora. U radu su ukratko opisani i već postojeći stvarni sustavi za detekciju umora koji se koriste u modernim automobilima.
Abstract (english)
In this paper, the system for drowsiness detection based on yawn analysis is described. The system takes upper lips and lower chin coordinates and calculates their distance. The calculated distance is then normalized and scaled in 0 to 1 interval. After that, the signal is brought to FIR filter where nonessential information is removed. From the filtered signal peak detector detects local maxima (signal peaks). The yawn detector takes these peaks, compares them to the threshold, and signals if yawn is detected. The timer takes signals of detected yawns and measures time between two yawn detection intervals. By using these intervals, yawn frequency is calculated. In the end, driver drowsiness is detected by setting yawn frequency threshold. The developed drowsiness detection is functional and can be used in real systems. However, this system is intended to be used as a part of a bigger one which uses more different information sources to make a valid decision. Besides practical part, in this paper the various scientific research papers which are focused on drowsiness detection and monitoring are shown. In these papers, the detection is based on EEG analysis, steering grip force analysis and face analysis. The EEG method is considered most accurate and it is used as a reference for others. The systems for drowsiness detection which are used in modern cars today are also described.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:079058
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 on2018-02-06 10:23:16