IEEE Consumer Electronics Magazine - October 2015 - 111

Attention
Tech.

Siemens
Supplies

Supplies

Smart Eye
Supplies

Supplies

Supplies Supplies Supplies

Bosch

Supplies
Supplies
Supplies
Supplies

Nissan

Supplies

Licenses
Saab/GM

AssistWare
Tech.

DCX

Delphi

Toyota

Denso

Collaborates

BASt
VW

Collaborates
Goverment
or Regulatory
Agency

Collaborates
Collaborates

Collaborate

University
of Michigan
Collaborates

OMC

Collaborates

Supplies
Collaborates

CMU

Collaborates

Seeing Supplies
NHTSA
Machines
Collaborates
BMW

Ford/Volvo

Collaborate

Visteon

University of
Pennsylvania

Collaborates

University of
Linköping

Scania

VTI
Academic
Institution

OEM

Tier-1
Supplier

Specialized
Entity

FIGURE 2. A sampling of companies, universities, and government offices involved in developing drowsiness/inattention detection
(applications and methods).

Drowsiness and fatigue are sometimes treated as equivalent
notions. However, they are different concepts. The word
fatigue "refers to the reluctance to continue a task as a result of
physical or mental exertion or a prolonged period of performing the same task" [3]. On the other hand, drowsiness or sleepiness could be explained as "the urge to fall sleep. It is the
result of a biological need to sleep that can be irresistible" [3].
Drowsiness can be induced or increased by factors such as
sleep disorders, medication, extended time spent on a task,
long periods of time awake, and inadequate or poor-quality
sleep [3], [32]. It can be inferred that drowsiness could be a
consequence of fatigue. Although the concepts are different,
some authors conclude that fatigue and drowsiness could be
considered similar because of their consequences [3], [59].
Attention is also a complicated concept related to how the person selects and processes information from the surrounding
world and generates a control action [53]. Attention usually
depends on a primary task or activity, which in this case would
be driving.
The term drowsiness could be equivalent to other definitions such as inattention, loss of vigilance, sleepiness, microsleep, tiredness, and distraction. This large quantity of
definitions shows the complexity of the term. Because of this
complexity, the features a DFD system has to screen and the
scale used to evaluate the severity of drowsiness are not easy
to comprehend, although some attempts have been made to
determine information related to periods of drowsiness [60].

Eye features provide various
advantages for a DFD system based
on machine-vision architectures.
Therefore, there are several features to review, including lane
departure, yawning, and blink rate.
Table 4 summarizes the different types of features used by
organizations developing DFD systems. Some of the most
common features (refer to Figure 1) are
▼ lane deviation, which is part of the "driving behavior" category
▼ steering-wheel movement, which is included in the "driving operation" category
▼ head-movement detection and mental reaction time, which
are physical reactions
▼ heart-rate variability (HRV), skin conductance, and electroencephalograms (EEGs), which are physiological reactions
▼ eye features [i.e., PERCLOS (the proportion of time in
1 min that the eyes are 80% closed), AVECLOS (the percentage of time the eyes are estimated to be fully closed over a
1-min interval), and blink rate], which are also categorized as
physiological reactions.
From reviewing, it can be concluded that several systems
use eye features (i.e., PERCLOS, AVECLOS, and blink rate)
to determine the driver's drowsiness level. Another fact is that
october 2015

IEEE Consumer Electronics Magazine

111

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - October 2015