Detecting driver drowsiness using feature-level fusion and user-specific classification

Jaeik Jo, Sung Joo Lee, Kang Ryoung Park, Ig Jae Kim, Jaihie Kim

Research output: Contribution to journalArticle

75 Citations (Scopus)

Abstract

Accurate classification of eye state is a prerequisite for preventing automobile accidents due to driver drowsiness. Previous methods of classification, based on features extracted for a single eye, are vulnerable to eye localization errors and visual obstructions, and most use a fixed threshold for classification, irrespective of variations in the driver's eye shape and texture. To address these deficiencies, we propose a new method for eye state classification that combines three innovations: (1) extraction and fusion of features from both eyes, (2) initialization of driver-specific thresholds to account for differences in eye shape and texture, and (3) modeling of driver-specific blinking patterns for normal (non-drowsy) driving. Experimental results show that the proposed method achieves significant improvements in detection accuracy.

Original languageEnglish
Pages (from-to)1139-1152
Number of pages14
JournalExpert Systems with Applications
Volume41
Issue number4 PART 1
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Fusion reactions
Textures
Automobiles
Accidents
Innovation

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

Jo, Jaeik ; Lee, Sung Joo ; Park, Kang Ryoung ; Kim, Ig Jae ; Kim, Jaihie. / Detecting driver drowsiness using feature-level fusion and user-specific classification. In: Expert Systems with Applications. 2014 ; Vol. 41, No. 4 PART 1. pp. 1139-1152.
@article{aa2770468d234d3fbf2d5c0b4a3192aa,
title = "Detecting driver drowsiness using feature-level fusion and user-specific classification",
abstract = "Accurate classification of eye state is a prerequisite for preventing automobile accidents due to driver drowsiness. Previous methods of classification, based on features extracted for a single eye, are vulnerable to eye localization errors and visual obstructions, and most use a fixed threshold for classification, irrespective of variations in the driver's eye shape and texture. To address these deficiencies, we propose a new method for eye state classification that combines three innovations: (1) extraction and fusion of features from both eyes, (2) initialization of driver-specific thresholds to account for differences in eye shape and texture, and (3) modeling of driver-specific blinking patterns for normal (non-drowsy) driving. Experimental results show that the proposed method achieves significant improvements in detection accuracy.",
author = "Jaeik Jo and Lee, {Sung Joo} and Park, {Kang Ryoung} and Kim, {Ig Jae} and Jaihie Kim",
year = "2014",
month = "1",
day = "1",
doi = "10.1016/j.eswa.2013.07.108",
language = "English",
volume = "41",
pages = "1139--1152",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",
number = "4 PART 1",

}

Detecting driver drowsiness using feature-level fusion and user-specific classification. / Jo, Jaeik; Lee, Sung Joo; Park, Kang Ryoung; Kim, Ig Jae; Kim, Jaihie.

In: Expert Systems with Applications, Vol. 41, No. 4 PART 1, 01.01.2014, p. 1139-1152.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Detecting driver drowsiness using feature-level fusion and user-specific classification

AU - Jo, Jaeik

AU - Lee, Sung Joo

AU - Park, Kang Ryoung

AU - Kim, Ig Jae

AU - Kim, Jaihie

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Accurate classification of eye state is a prerequisite for preventing automobile accidents due to driver drowsiness. Previous methods of classification, based on features extracted for a single eye, are vulnerable to eye localization errors and visual obstructions, and most use a fixed threshold for classification, irrespective of variations in the driver's eye shape and texture. To address these deficiencies, we propose a new method for eye state classification that combines three innovations: (1) extraction and fusion of features from both eyes, (2) initialization of driver-specific thresholds to account for differences in eye shape and texture, and (3) modeling of driver-specific blinking patterns for normal (non-drowsy) driving. Experimental results show that the proposed method achieves significant improvements in detection accuracy.

AB - Accurate classification of eye state is a prerequisite for preventing automobile accidents due to driver drowsiness. Previous methods of classification, based on features extracted for a single eye, are vulnerable to eye localization errors and visual obstructions, and most use a fixed threshold for classification, irrespective of variations in the driver's eye shape and texture. To address these deficiencies, we propose a new method for eye state classification that combines three innovations: (1) extraction and fusion of features from both eyes, (2) initialization of driver-specific thresholds to account for differences in eye shape and texture, and (3) modeling of driver-specific blinking patterns for normal (non-drowsy) driving. Experimental results show that the proposed method achieves significant improvements in detection accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84888363097&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888363097&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2013.07.108

DO - 10.1016/j.eswa.2013.07.108

M3 - Article

AN - SCOPUS:84888363097

VL - 41

SP - 1139

EP - 1152

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 4 PART 1

ER -