Classifiers for driver activity monitoring

Harini Veeraraghavan, Nathaniel Bird, Stefan Atev, Nikolaos P Papanikolopoulos

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

The goal of this work is the detection and classification of driver activities in an automobile using computer vision. To this end, this paper presents a novel two-step classification algorithm, namely, an unsupervised clustering algorithm for grouping the actions of a driver during a certain period of time, followed by a supervised activity classification algorithm. The main contribution of this work is the combination of the two methods to provide a computationally fast solution for deployment in real-world scenarios that is robust to illumination and segmentation issues under most conditions experienced in the automobile environment. The unsupervised clustering groups the actions of the driver based on the relative motion detected using a skin-color segmentation algorithm, while the activity classifier is a binary Bayesian eigenimage classifier. Activities are grouped as safe or unsafe and the results of the classification are shown on several subjects obtained from two distinct driving video sequences.

Original languageEnglish (US)
Pages (from-to)51-67
Number of pages17
JournalTransportation Research Part C: Emerging Technologies
Volume15
Issue number1
DOIs
StatePublished - Jan 1 2007

Fingerprint

Classifiers
driver
monitoring
Monitoring
Automobiles
motor vehicle
Clustering algorithms
grouping
Computer vision
Skin
video
Lighting
scenario
Color
Classifier
Group
segmentation
Automobile
Segmentation

Keywords

  • Activity classification
  • Activity history images
  • Driver activity
  • Eigenimage classification
  • Unsupervised clustering

Cite this

Classifiers for driver activity monitoring. / Veeraraghavan, Harini; Bird, Nathaniel; Atev, Stefan; Papanikolopoulos, Nikolaos P.

In: Transportation Research Part C: Emerging Technologies, Vol. 15, No. 1, 01.01.2007, p. 51-67.

Research output: Contribution to journalArticle

Veeraraghavan, Harini ; Bird, Nathaniel ; Atev, Stefan ; Papanikolopoulos, Nikolaos P. / Classifiers for driver activity monitoring. In: Transportation Research Part C: Emerging Technologies. 2007 ; Vol. 15, No. 1. pp. 51-67.
@article{5e431ee8c96e4f67ba29f9b0fde7d6e5,
title = "Classifiers for driver activity monitoring",
abstract = "The goal of this work is the detection and classification of driver activities in an automobile using computer vision. To this end, this paper presents a novel two-step classification algorithm, namely, an unsupervised clustering algorithm for grouping the actions of a driver during a certain period of time, followed by a supervised activity classification algorithm. The main contribution of this work is the combination of the two methods to provide a computationally fast solution for deployment in real-world scenarios that is robust to illumination and segmentation issues under most conditions experienced in the automobile environment. The unsupervised clustering groups the actions of the driver based on the relative motion detected using a skin-color segmentation algorithm, while the activity classifier is a binary Bayesian eigenimage classifier. Activities are grouped as safe or unsafe and the results of the classification are shown on several subjects obtained from two distinct driving video sequences.",
keywords = "Activity classification, Activity history images, Driver activity, Eigenimage classification, Unsupervised clustering",
author = "Harini Veeraraghavan and Nathaniel Bird and Stefan Atev and Papanikolopoulos, {Nikolaos P}",
year = "2007",
month = "1",
day = "1",
doi = "10.1016/j.trc.2007.01.001",
language = "English (US)",
volume = "15",
pages = "51--67",
journal = "Transportation Research Part C: Emerging Technologies",
issn = "0968-090X",
publisher = "Elsevier Limited",
number = "1",

}

TY - JOUR

T1 - Classifiers for driver activity monitoring

AU - Veeraraghavan, Harini

AU - Bird, Nathaniel

AU - Atev, Stefan

AU - Papanikolopoulos, Nikolaos P

PY - 2007/1/1

Y1 - 2007/1/1

N2 - The goal of this work is the detection and classification of driver activities in an automobile using computer vision. To this end, this paper presents a novel two-step classification algorithm, namely, an unsupervised clustering algorithm for grouping the actions of a driver during a certain period of time, followed by a supervised activity classification algorithm. The main contribution of this work is the combination of the two methods to provide a computationally fast solution for deployment in real-world scenarios that is robust to illumination and segmentation issues under most conditions experienced in the automobile environment. The unsupervised clustering groups the actions of the driver based on the relative motion detected using a skin-color segmentation algorithm, while the activity classifier is a binary Bayesian eigenimage classifier. Activities are grouped as safe or unsafe and the results of the classification are shown on several subjects obtained from two distinct driving video sequences.

AB - The goal of this work is the detection and classification of driver activities in an automobile using computer vision. To this end, this paper presents a novel two-step classification algorithm, namely, an unsupervised clustering algorithm for grouping the actions of a driver during a certain period of time, followed by a supervised activity classification algorithm. The main contribution of this work is the combination of the two methods to provide a computationally fast solution for deployment in real-world scenarios that is robust to illumination and segmentation issues under most conditions experienced in the automobile environment. The unsupervised clustering groups the actions of the driver based on the relative motion detected using a skin-color segmentation algorithm, while the activity classifier is a binary Bayesian eigenimage classifier. Activities are grouped as safe or unsafe and the results of the classification are shown on several subjects obtained from two distinct driving video sequences.

KW - Activity classification

KW - Activity history images

KW - Driver activity

KW - Eigenimage classification

KW - Unsupervised clustering

UR - http://www.scopus.com/inward/record.url?scp=33947316643&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947316643&partnerID=8YFLogxK

U2 - 10.1016/j.trc.2007.01.001

DO - 10.1016/j.trc.2007.01.001

M3 - Article

VL - 15

SP - 51

EP - 67

JO - Transportation Research Part C: Emerging Technologies

JF - Transportation Research Part C: Emerging Technologies

SN - 0968-090X

IS - 1

ER -