Abstract
Traffic accidents due to human
error causes many death and injuries around in the world. Driver drowsiness is
leading causes of motor vehicle crashes. The sensation of drowsiness diminishes
the level of vigilance of the driver and results in perilous situations. While
eminent automobile manufacturers like Volvo, Mercedes-Benz and Bosch have
ventured into the development of drowsiness detection technologies, use of
these safety systems is not widespread among drivers due to their availability
in luxury cars only.
The Goal of this thesis project
proposes a non-intrusive Real-Time Driver Drowsiness Detection System using
Deep Learning and Raspberry Pi to utilize the driving video dataset developed
by the University of Texas, Arlington for training and testing the deep
learning model. The developed model will then be deployed on Raspberry Pi (with
Pi Camera module) to make predictions and alert the driver in real time. Road
crashes and related forms of accident are common cause of injury and
death. The device is currently essential in many fields for sleepiness related
accident prevention. Real-time driver drowsiness system alerts users when they
are fall in a sleep. Real Time Driver Drowsiness Detection system is the
complete system is implemented on Raspberry Pi which uses a webcam to monitor
user’s facial Expression and average Face duration to detect drowsiness. The
project purpose to a car safety which help to prevent accidents cause by the driver
get to drowsy. The motive of this project is to
detect multi-stage drowsiness target not only extreme and easily visible case
but also subtitle cases when micro-expressions are the discriminative factors.
Detection of these subtle cases can be important for detecting drowsiness at an
early stage, so as to activate drowsiness prevention mechanisms. The project
aims to build a computationally inexpensive deep learning model that can be
deployed on Raspberry Pi. The model developed will receive video input from the
Pi Camera module placed on the vehicle dashboard. It will use this video
stream to gauge the alertness of the driver and notify the driver when early
stages of drowsiness are detected.
Keyword:-Machine
learning, Driver observance System; temporary state Detection; Deep Learning;
Raspberry Pi., Android, Neural network.
ACKNOWLDGEMENT
I am extremely fortunate to be involved in an
exciting and challenging research project “Study
of Real Time Driver Drowsiness Detection”.it has enriched my life, giving
me an opportunity to work in a new environment of Research area. This project
increased my thinking and understanding capability and after the completion of
this project ,I experience the feeling of achievement and satisfaction.
I would like to express my greatest gratitude and
respect to my supervisor Assistant
professor Er. Pyuish rai and co-supervisor Er.Nidhi Prasad,for his excellent guidance , valuable suggestions
and endless support .They have not only be able to wok under guidance of such
dynamic personalities.
I express my sincere thanks to Er.pyuish rai ,Er.Awdhesh Dixit and Er. Shobhit srivastava Assistant professor, M.Tech, Department of
computer science &Engineering, IET Ayodhya for their full-time support and
motivation of research and development in my M.Tech program I am also grateful
to Prof. Manoj Dixit (Hummable
Vice-Chancellor), Prof. Rama Patti
Mishra (Director) Err. Hashish kumar
pandey (HOD), All faculty members and staff of IET, Dr. Ram manohar Lohia
Avadh University, Ayodhya for providing all the facilities and support .
It is a pleasure to Acknowldge the support and help
extended by all my colleagues. Last but not the least; I want to convey my
Heartiest gratitude to my parents for their immeasurable love, support and
encouragement.
Chanchal singh
181104
M.tech (computer science Engineering)
Get Free Quote!
271 Experts Online