Drowsiness Alert Using Machine Learning Algorithms and Deep Learning Algorithms

: We proposed to use this approach to reduce the number of accidents caused by driver weariness, thereby improving road safety. This device uses visual data and artificial intelligence to identify driver drowsiness automatically. For the neural transfer function, we use SoftMax to find, monitor, and examine both the driver's face and eyes in order to measure PERCLOS (% of eye closure). It will also employ alcohol pulse detection to evaluate whether or not the user is normal. Due to long driving periods and boredom in overcrowded conditions, driver fatigue is one of the primary causes of traffic accidents, particularly for drivers of large vehicles (such as buses and heavy trucks).


I. INTRODUCTION
There are two types of methodologies for detecting driver drowsiness: behavior-based approaches and eye and neck angle-based approaches.Recent studies show that algorithms that calculate the distance between the driver's open and closed eyes have more reliability and accuracy than other methods for detecting driver drowsiness The proposed system makes use of parameter conduct:  Eye blinking, yawning, eye openness, jaw angle, as well as other behavior parameters are included in the personality parameter. A camera mounted on the bus captures the live video. The video is separated into frames, after which the photos are selected from the frames. Image noise is reduced by taking individual images.

Motivation
Monitoring the driver's actions while driving by examining the vehicle's maneuverability may be a very essential task in terms of improving driving safety.The most crucial factor to be identified will be the ability to distinguish between unintended and deliberate automobile steering wheel inputs, such as a sudden significant steering input could indicate the driver's level of alertness.

II. LITERATURE REVIEW Paper Name: Towards Detection of Bus Driver Fatigue Based on Robust Visual Analysis of Eye State
Author: Bappaditya Mandal, Liyuan Li, Gang Sam Wang, and Jie Lin Abstract: Because of extended driving times and boredom in the workplace, driver weariness is one of the leading causes of traffic accidents, especially among big vehicle drivers (such as buses and heavy trucks).We present a vision-based tiredness detection system for bus driver monitoring in this study, which is simple and adaptable enough to be used in buses and big vehicles.Head-shoulder detection, face detection, eye detection, eye openness estimation, fusion, sleepiness measure percentage of eyelid closure (PERCLOS) estimation, and tiredness level classification are among the modules included in the system.The following are the most important innovative techniques: 1) a method based on spectral regression for estimating the continuous amount of eye opening.

III. DESIGN AND ANALYSIS
Figure 1: System Architecture The system which detects driver sleepiness starts monitoring the driver's steering behaviour as soon as the trip starts.It then detects changes over the duration of long excursions, as well as the driver's tiredness state.

IV. PROPOSED ALGORITHMS 4.1 Convolutional Neural Network (CNN)
CNN is a Deep Learning algorithm that can take an input image, assign priority to various aspects/objects in the image, and distinguish between them.A CNN, or convolutional neural network, is a deep learning neural network designed to analyse organised arrays of data like representations.

Haar Cascade
This is an Object Detection Algorithm that is used to recognise faces in images or real-time videos.Edge or line are used in the algorithm.A Haar-like feature averages the pixel intensities in neighbouring rectangular sections at a specified place in a detection window and calculates the difference between these sums.This distinction is then utilised to categorise picture subsections.
V. IMPLEMENTATION 5.1 Yawn Detection On different resolutions images of faces and eyes taken from an oblique viewing angle, our proposed approach can discriminate between simulated drowsy and sleepy states and the typical condition of driving.As a result, our technology may be able to effectively monitor the attention level of bus drivers without the need for additional cameras.Existing visionbased strategies for detecting driver fatigue may be able to improve their capabilities and application using our approach.
Paper Name: Portable Prevention and Monitoring of Driver's Drowsiness Focuses to Eyelid Movement Using Internet of Things Author: Menchie Miranda,Alonic a Villanue va; Mar k Jomar Buo; Re ynald Abstract: Since the number of vehicular accidents in the Philippines has lately increased, this article offers a sleepiness prevention device.To raise driver awareness, current safety measures are used, such as the usage of standard rumble strips on roadways, GPS installation, and speed limiters.

Figure 2 :
Figure 2: Current Probability of Yawn