International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 05 | May 2022
p-ISSN: 2395-0072
www.irjet.net
Driver Drowsiness Detection System Using Image Processing Saily Joshi, Anirudh Nair, Rasika Dongare Saily Joshi, 411045, India Anirudh Nair, Pune, 411061, India Rasika Dongare, Pune, 411012, India Under the guidance of: Prof. Swati Kale, Dept. of E&TC Engineering, JSPM’s RSCOE, Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - We propose to diminish the number of disasters
measurement. However, biological parameter measurement is part invasive in nature requiring probes and direct physical measurement which has limited practicality. The advantage of monitoring the driver’s eye response is that there is no probe or physical contact required with the driver besides it being completely non-disturbing kind of a measurement. Absence of any interference, disturbance, attachment of probes and ease of implementation makes monitoring of eyes for closure patterns the best tool to detect sleepiness of a driver. The data of eye closure gathered can be used in an algorithm that can correlate the pattern of eye closure to detect the sleepiness or otherwise of a driver during active driving phases or also during rest intervals.
realized by driver weariness and along these lines improve road prosperity. This structure treats the modified disclosure of driver sluggishness subject to visual information what’s more, man-made awareness. We discover, track and explore both the driver face and eyes to measure PERCLOS (level of eye end) with Softmax for neural trade work. It will be in like manner uses liqour and beat acknowledgement to take a gander at the individual is average or odd. Driver’s exhaustion is one of the huge purposes behind car crashes, particularly for drivers of gigantic vehicles (for instance, transports and overpowering trucks) due to deferred driving periods and weakness in included conditions. This method can be inferred as a low cost and reliable way to minimize the number of injuries related to the driver’s drowsiness in order to improve the health of travel.
2. RELATED WORK There has been a lot of development recently in the area of smart vehicle software. Due to the driver sleepiness by exhaustion, which also results in frequent accidents, research is being done in the area. In some research papers, an ensemble deep learning architecture has been proposed which not only looks at the eyes and mouth samples of the driver but also includes the fitness of the driver in the decision structure. [1] There are also some works in which eye retina detection and facial feature extraction is used. [2] Prevention of driver drowsiness is the main area of focus of all of these works.
Key Words: Raspberry Pi, Eye tracking, Driver, Image Processing.
1. INTRODUCTION The tendency of drivers to fall asleep is one of the main causes of road accidents. The data from various countries shows that around 20 percent of road injuries are a result of road accidents which are caused by the drivers falling asleep while driving. It is evident that the number of fatalities in road accidents is in a great way contributed by this hazard of drivers falling asleep when driving. Consumption of alcohol, fatigue during driving and a casual approach to driving are the main contributors to this risky behavior. It is due to this that, several people across various nations are adversely impacted. It is monitoring of drivers for drowsiness that will be an important tool in training and controlling the behavior of drivers to fall asleep when driving. It is this monitoring tool that will give an early warning of the tendency of a driver to fall asleep and prevent an accident. This paper presents a project for monitoring the sleepiness of a driver that is liable to cause an accident. The technology used is Open CV, Raspberry Pi and Image Processing. There is imperial evidence to show that there are various possibilities that can be implemented to detect sleepiness within the drivers. The measurement of sleepiness can be done by using eye response, vital biological parameters and the effectiveness of driving itself. Of these, eye response and biological monitoring are more reliable forms of
© 2022, IRJET
|
Impact Factor value: 7.529
The number one causes for driver sleepiness are fatigue, consumption of alcohol, all of which can be prevented by the use of technology, especially artificial intelligence. Furthermore, artificial intelligence has also been previously used to detect the rate at which the driver is becoming drowsy.[3] There have also been more invasive ways of preventing driver sleepiness, which result in a more uncomfortable driving experience, hence the need for the proposed system. EEG has been used for the driver sleepiness detection. Here, raw EEG is taken and multiple functions are performed on it. [4]
3. PROPOSED METHODOLOGY Given below is a flowchart that can quickly analyze the information to detect sleepiness of a driver. The number of
|
ISO 9001:2008 Certified Journal
|
Page 3676