Online driver distraction detection using long short-term memory

Martin Wollmer; Christoph Blaschke; Thomas Schindl; Bjrn Schuller; Berthold Farber; Stefan Mayer; Benjamin Trefflich

doi:10.1109/TITS.2011.2119483

Online driver distraction detection using long short-term memory

Martin Wollmer, Christoph Blaschke, Thomas Schindl, Bjrn Schuller, Berthold Farber, Stefan Mayer, Benjamin Trefflich

Chair of Health Informatics

Research output: Contribution to journal › Article › peer-review

159 Scopus citations

Abstract

Lane-keeping assistance systems for vehicles may be more acceptable to users if the assistance was adaptive to the driver's state. To adapt systems in this way, a method for detection of driver distraction is needed. Thus, we propose a novel technique for online detection of driver's distraction, modeling the long-range temporal context of driving and head tracking data. We show that long short-term memory (LSTM) recurrent neural networks enable a reliable subject-independent detection of inattention with an accuracy of up to 96.6%. Thereby, our LSTM framework significantly outperforms conventional approaches such as support vector machines (SVMs).

Original language	English
Article number	5732698
Pages (from-to)	574-582
Number of pages	9
Journal	IEEE Transactions on Intelligent Transportation Systems
Volume	12
Issue number	2
DOIs	https://doi.org/10.1109/TITS.2011.2119483
State	Published - Jun 2011

Keywords

Driver assistance systems
driver state estimation
long short-term memory (LSTM)
recurrent neural networks (RNNs)

Access to Document

10.1109/TITS.2011.2119483

Cite this

@article{c59d3054d4b84b87bb4eec2b24d9dcf8,

title = "Online driver distraction detection using long short-term memory",

abstract = "Lane-keeping assistance systems for vehicles may be more acceptable to users if the assistance was adaptive to the driver's state. To adapt systems in this way, a method for detection of driver distraction is needed. Thus, we propose a novel technique for online detection of driver's distraction, modeling the long-range temporal context of driving and head tracking data. We show that long short-term memory (LSTM) recurrent neural networks enable a reliable subject-independent detection of inattention with an accuracy of up to 96.6%. Thereby, our LSTM framework significantly outperforms conventional approaches such as support vector machines (SVMs).",

keywords = "Driver assistance systems, driver state estimation, long short-term memory (LSTM), recurrent neural networks (RNNs)",

author = "Martin Wollmer and Christoph Blaschke and Thomas Schindl and Bjrn Schuller and Berthold Farber and Stefan Mayer and Benjamin Trefflich",

year = "2011",

month = jun,

doi = "10.1109/TITS.2011.2119483",

language = "English",

volume = "12",

pages = "574--582",

journal = "IEEE Transactions on Intelligent Transportation Systems",

issn = "1524-9050",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - Online driver distraction detection using long short-term memory

AU - Wollmer, Martin

AU - Blaschke, Christoph

AU - Schindl, Thomas

AU - Schuller, Bjrn

AU - Farber, Berthold

AU - Mayer, Stefan

AU - Trefflich, Benjamin

PY - 2011/6

Y1 - 2011/6

N2 - Lane-keeping assistance systems for vehicles may be more acceptable to users if the assistance was adaptive to the driver's state. To adapt systems in this way, a method for detection of driver distraction is needed. Thus, we propose a novel technique for online detection of driver's distraction, modeling the long-range temporal context of driving and head tracking data. We show that long short-term memory (LSTM) recurrent neural networks enable a reliable subject-independent detection of inattention with an accuracy of up to 96.6%. Thereby, our LSTM framework significantly outperforms conventional approaches such as support vector machines (SVMs).

AB - Lane-keeping assistance systems for vehicles may be more acceptable to users if the assistance was adaptive to the driver's state. To adapt systems in this way, a method for detection of driver distraction is needed. Thus, we propose a novel technique for online detection of driver's distraction, modeling the long-range temporal context of driving and head tracking data. We show that long short-term memory (LSTM) recurrent neural networks enable a reliable subject-independent detection of inattention with an accuracy of up to 96.6%. Thereby, our LSTM framework significantly outperforms conventional approaches such as support vector machines (SVMs).

KW - Driver assistance systems

KW - driver state estimation

KW - long short-term memory (LSTM)

KW - recurrent neural networks (RNNs)

UR - http://www.scopus.com/inward/record.url?scp=79958176949&partnerID=8YFLogxK

U2 - 10.1109/TITS.2011.2119483

DO - 10.1109/TITS.2011.2119483

M3 - Article

AN - SCOPUS:79958176949

SN - 1524-9050

VL - 12

SP - 574

EP - 582

JO - IEEE Transactions on Intelligent Transportation Systems

JF - IEEE Transactions on Intelligent Transportation Systems

IS - 2

M1 - 5732698

ER -

Online driver distraction detection using long short-term memory

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this