CLASSIFICATION OF DRIVER'S DROWSINESS FROM STEERING WHEEL MOTION UNDER REAL TRAFFIC CONDITIONS

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F12%3APU99587" target="_blank" >RIV/00216305:26220/12:PU99587 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

CLASSIFICATION OF DRIVER'S DROWSINESS FROM STEERING WHEEL MOTION UNDER REAL TRAFFIC CONDITIONS

Popis výsledku v původním jazyce

To develop a system for drivers drowsiness recognition is a challenging task in the modern car transportation. Many studies have promising results. Unfortunately most data is acquired in the laboratory conditions. Therefore proving drowsiness detection reliability and accuracy in real traffic is difficult. The analyzed data in this paper is acquired from the real traffic and hence it contains all uncertainty. An in-direct measurement from the vehicle CAN bus has been chosen for data acquisition in orderto not affect the driver. The data is preprocessed according to the assumptions about drivers behavior and transformed to the frequency domain by means of the orthogonal transform (STFT, CWT and DWT). Subsequently, in the frequency domain, more than 70000 features are generated. By means of the filter feature selection, 10 best features are chosen for a prediction. Finally, 1-NN model is used for prediction accuracy estimation.

Název v anglickém jazyce

CLASSIFICATION OF DRIVER'S DROWSINESS FROM STEERING WHEEL MOTION UNDER REAL TRAFFIC CONDITIONS

Popis výsledku anglicky

To develop a system for drivers drowsiness recognition is a challenging task in the modern car transportation. Many studies have promising results. Unfortunately most data is acquired in the laboratory conditions. Therefore proving drowsiness detection reliability and accuracy in real traffic is difficult. The analyzed data in this paper is acquired from the real traffic and hence it contains all uncertainty. An in-direct measurement from the vehicle CAN bus has been chosen for data acquisition in orderto not affect the driver. The data is preprocessed according to the assumptions about drivers behavior and transformed to the frequency domain by means of the orthogonal transform (STFT, CWT and DWT). Subsequently, in the frequency domain, more than 70000 features are generated. By means of the filter feature selection, 10 best features are chosen for a prediction. Finally, 1-NN model is used for prediction accuracy estimation.

Druh

D - Stať ve sborníku

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

<a href="/cs/project/GA102%2F09%2F1897" target="_blank" >GA102/09/1897: Bezpečnost automobilové dopravy - BAD</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2012

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

MENDEL 2012, 18th International Conference on Soft Computing

ISBN

978-80-214-4540-6

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

428-433

Název nakladatele

Neuveden

Místo vydání

Neuveden

Místo konání akce

Brno University of Technology

Datum konání akce

27. 6. 2012

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—