Driver's Drowsiness Detection and Alcohol Detection System With Auto Engine Lock

Driver’s Drowsiness Detection and Alcohol Detection System with Auto
Engine Lock
Abstract
Road accidents are being major reason for fatalities in recent times. Accidents not only cause
loss of life but also causing implications road safety turned into major challenge in automobile
industry. Among various reasons for road accidents drunk and drive and driver drowsiness are
taking major proportion, continuous monitoring of driver’s alertness will help in minimizing
the crashes. This project aims to measure alcohol content and drowsiness by capturing image
continuously and measuring the state of the eye according to the desired algorithm and alcohol
sensor (MQ3 Sensor) is used to detect the alcohol consumption of the driver so that the risk of
accidents can be reduced. Though there are several methods for measuring the drowsiness but
this approach is completely non-intrusive which does not affect the driver in any way, hence
giving the exact condition of the driver.
For detection of drowsiness, EAR (Eye Aspect Ratio) is considered. So when the EAR exceeds
a certain amount then the driver is identified to be sleepy. For implementing this system
several OpenCV libraries are used including NUMPY. alcohol sensor is used to detect the
alcohol consumption of the driver so that the risk of accidents can be reduced. The entire
system is implemented using Raspberry-Pi.
Keywords: MQ3 Sensor, EAR, OpenCV, NUMPY, Raspberry Pi, etc.
Introduction
Driving while drowsy or in drunk state are the two main reasons for traffic accidents and its
related financial losses. Most of countries cost 3% of their GDP due to road crashes and nearly
about 1.35 million people die each year. In case of India, India constitutes 1% of vehicle
population but 6% of total global road accidents and approximately 151 thousand road accident
deaths. One of the primaries agencies that conducts research on road and driver safety is the
national highway traffic safety administration. In study looking at critical reasons for road
accidents 94% of car accidents are caused by drivers out of which 25 to 30% of accidents are
caused due to drowsiness and drunk and drive. Apart from improvised automobile structures to
reduce the accidents, drivers’ attitude while driving is not addressed properly to evaluate drivers
in attention, there are three main approaches to go through they are visual based approach,
driving behavior-based approach and physiological technique involving analysis of vital signals
like brain activity, heart and pulse rate. This requires the usage of electronic devices which are
attached to drivers’ body can act as annoyance to the driver.
Driving behavior-based approach evaluates drivers’ behavior over period of time by
observing variations in the speed, steering wheel angle, acceleration, lateral position and
breaking.
Through this parameter driver’s attention and activities can be measured. liang-et al
demonstrated a real-time approach to detect driver’s distraction using eye movements and
driving performance, and date was made to train and test both support vector machining[svm]
and logistics regression models[lrm] to recognize driver’s distraction
Visual based approach analyses drivers’ facial images like eye blinks, yawning and head
movements. Hammond et.al proposed drowsiness detection system through assessment of eye
status in the near infrared spectrum. Moriyama et al predicted the eye status by using templates
to match with eye lids. Number of eye blinks of the driver is another method to detect
drowsiness. This approach based on visual features are preferred because of its non-interference
with the driver.
Objectives of Project
Road accidents due to drunk and driving are the most important challenges in safety system of
automobile industry. The objective of this project is to detect the drowsiness of the driver using
visual feature approach by analyzing the eye and face movements. Alcohol sensor is used to
detect the drunkenness of the driver. IF the driver found in in alertness or inactive state, alarm
would be generated via buzzer and using a vibrator to alert the driver. If the alert would go off
for more than three times within the specified time period, then engine would go off
immediately.
Scope of Project
The proposed system is used to avoid various road accidents caused by drowsy driving and
drunken driving.
This system is used for security purpose of a driver to caution the driver if any fire accident or
any gas leakage occur.
The system is used to detect the alcohol consumption of the driver in order to reduce rash
driving.
Drowsy driver detection methods can form the basis of a system to potentially reduce the
number of crashes related to drowsy driving.
Literature Review
Drowsiness is synonymous with sleepy that means an inclination to fall asleep. The stages of
sleep are often classified as awake, non-rapid eye movement sleep and rapid eye movement
sleep. The second stage NREM, is often into subsequent 3 stages:
stage 1: Transition from awake to asleep(drowsy)
stage 2: lightweight sleep
stage 3: deep sleep
Drowsiness cannot be manifested during real time environment and needs to be developed and
tested in a laboratory setting. In a laboratory setting, the foremost reliable and informative
information that pertains to driver drowsiness depends only on the approach in which the driver
falls into the drowsy state. Drowsiness depends on the quality of last night sleep and the
biological time.
Peters-et-al studies an equivalent subject and regarded results of no sleep deprivation, partial
sleep deprivation and total sleep deprivation on their drowsiness level. Even in partial sleep
deprivation, person tends to drowse after a while.
otami.et al found that sleep deprivation is not only the reasons for the drowsiness but also
prolonged driving can cause the drowsiness. It has been inferred that subjects tends to become
drowsy at intervals of twenty to twenty-five min of driving.
Drowsiness can be detected using different techniques. Template match making is one of
method. In which two pictures are compared to get the deviation. To recognize the matching
territory picture can be contrasted against the format picture by sliding. sliding is moving the
patch one pixel at once. Brightest area indicates the highest matches.
Principal component analysis and linear discriminate analysis are used for blink detection
considering driver’s tiredness and diversion HARR cascade classifier method is used for facial
eye detection methods utilizing the real time image processing.
Physiological measures are also used to detect the drowsiness. Heat rate varies considerably
between different stages of drowsiness and fatigue. Therefore, heart rate determined by ECG
signal can even be used to observe drowsiness. But still drowsiness detection through
behavioral attributes and established better results than using physiological measures.
This project acquaints the vision-based strategy with distinguished drowsiness. significant
difficulties are face recognition and iris location under different conditions and creating real
time systems
Block Diagram
Camer
LCD
SD Card
Raspberry L298N Motor
MQ 2 (Smoke Pi
Controller Relay
MQ 3 (Alcohol Sensor)
Buzzer
HDMI to VGA
Power
Fig 1. System Architecture
Conclusion
A system can be developed to monitor the drowsiness and alcohol detection. It is non-intrusive
in nature thus do not cause any annoyance to the driver. Information about the head and eye
positions will be obtained by using many self-determined algorithms. When the eyes are closed
exceeding certain period of time then automatic alert will be sent. In case of extreme conditions
engine will be switched off. We will use image processing which achieves high accuracy and
reliable detection of drowsiness. When coming to alcohol consumption, buzzer will sound
immediately if anyone with alcohol consumption enters into vehicle. Thus we will address the
major causes for the fatal crashes of vehicles and embedded kit for drowsiness detection will be
developed and will be placed in vehicle.
References
[1] S. G. Klauer, T. A. Dingus, V. L. Neale, Sudweeks, and Ramsey, “The impact of driver
inattention on nearcrash/crash risk: An analysis using the 100-car naturalistic driving study
data,” Tech. Rep., 2006.
[2] J. F. Coughlin, B. Reimer, and B. Mehler, “Monitoring, managing, and Motivating driver
safety and wellbeing.” IEEE Pervasive Computing, vol. 10, no.3, 2011.
[3] T. Yoshioka, S. Nakashima, J. Odagiri, H. Tomimori, and T. Fukui, “Pupil detection in the
presence of specular reflection,” in Proceedings of the Symposium on Eye Tracking Research
and Applications. ACM, 2014, pp. 363–364.
[4] L. ´ Swirski, A. Bulling, and N. Dodgson, “Robust real-time pupil tracking in highly off-
axis images,” in Proceedings of the Symposium on Eye Tracking Research and Applications.
ACM, 2012, pp. 173–176.
[5] M. Muoz, J. Lee, B. Reimer, B. Mehler, and T. Victor, “Analysis of drivers’head and eye
movement correspondence: Predicting drivers’ glance location using head rotation data,” in
Proceedings of the 8th International Driving Symposium on Human Factors in Driver
Assessment, Training, and Vehicle Design, Snowbird,UT, 2015, to Appear.
[6] W. J. Talamonti, W. Huang, L. Tijerina, and D. Kochhar, “Eye glance and head turn
correspondence during secondary task performance in simulator driving,” in Proceedings of
the Human Factors and Ergonomics Society Annual Meeting, vol.57, no. 1. SAGE
Publications, 2013, pp. 1968–1972

Driver's Drowsiness Detection and Alcohol Detection System With Auto Engine Lock

Uploaded by

Copyright:

Available Formats

Driver's Drowsiness Detection and Alcohol Detection System With Auto Engine Lock

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Driver's Drowsiness Detection and Alcohol Detection System With Auto Engine Lock

Uploaded by

Copyright:

Available Formats

Driver’s Drowsiness Detection and Alcohol Detection System with Auto

Keywords: MQ3 Sensor, EAR, OpenCV, NUMPY, Raspberry Pi, etc.

stage 1: Transition from awake to asleep(drowsy)

stage 2: lightweight sleep

stage 3: deep sleep

Fig 1. System Architecture

You might also like