A Review on Drowsy Driver Detection in Digital Image Processing

Abstract

Somnolence or drowsiness is a condition of strong urge for rest, or sleeping at unusual time or instantly while indulging with some task (like hypersomnia). It has unmistakable implications and causes. It can allude to the typical state going before falling asleep or a manifestation of other medical conditions. Drowsiness can be hazardous when performing errands that require steady focus, like driving a vehicle. At the point when an individual is adequately exhausted, microsleeps might be capable. In people denied of rest, somnolence may immediately disperse for short timeframes; this marvel is the revitalizing burst of energy, and results from the normal cycling of the circadian musicality meddling with the cycles the body does to set itself up to rest. Languid driving is a major issue in the India. The danger, risk, and regularly terrible consequences of lazy driving are disturbing. Languid driving is the hazardous blend of driving and languor or weariness. In this paper, a survey has been done about the different identification strategies for driver\\\'s drowsiness detection. There are several researches have been done till now but somewhere somehow they suffers due to accuracy especially at real time.

Introduction

I. INTRODUCTION

Driver drowsiness has been the key issue for endless occurrences on account of sluggishness, long road condition, and irksome climate [1]. Consistently, the National Highway Traffic Safety Administration (NHTSA) and World Health Organization (WHO) have detailed that roughly 1.35 million individuals bite the dust because of vehicle crashes across the world. All around, accidents happen due to lacking technique for driving [2].

These conditions arise on the off chance that the driver is dependent on alcohol or in drowsiness [3]. The most outrageous kinds of destructive incidents are seen as a serious component of sluggishness of the driver. Right when drivers fall asleep, the command over the vehicle is lost [4].

There is a need to setup splendid or savvy vehicle system through different methods [5]. Sixteen Indians kicked the bucket in street setbacks consistently. As demonstrated by the Global Road Safety Report 2015, all out 141,526 people were killed and around five lakh people hurt in India because of street mishaps.

Anyway this number isn't true to form evaluated in light of the fact that all setbacks are not paid all due respects to the police. In the US reliably around 100,000 mishaps happens in view of driver drowsiness or fatigue surveyed by National Highway Traffic Safety Administration (NHTSA).

National Highway Traffic Safety Administration uncovered that in 2013 by virtue of driver drowsiness 72,000 mishaps, 44,000 injuries and 800 passings occurred. These days, every individual purposes a vehicle. It is ordinary considered as luxury yet it has now turned into a need in a normal individual's life.

People are particularly stressed over their prosperity and moreover the vehicles security is it on the off chance that there ought to emerge an event of burglary or in case of a setback. Motor vehicles are utilized for certain, reasons like for transportation in general populace, things and moreover confidential trips. During long driving hours the driver gets depleted and the singular tendency drowsy yet keeps driving for showing up at the endpoint early. The driving in non-safe condition due to driver will in everyday undertaking. Right when the singular feels lethargic the individual being referred to really keeps driving ignoring how it is hazardous. The person being referred to falls asleep and the vehicle isn't anything else in control and bangs into various vehicles on street provoking loss of various lives.   

Driver help structure progression have been expected to prevent the disasters as a result of driver drowsiness, since all the time he can't deal with the vehicles a couple of risks may happed due to driver's laziness, or carelessness. This system helps with bringing the thought of a driver. Ordinarily the result of drowsiness and interference is while driving unreasonable to give the thought on driving it is related to the degree of non obsession. Driver falling asleep the lazy driver recognizable proof system recognizes the drained and drowsiness of the driver and it offers forewarning to the driver. There are various advances are open to recognize the driver drowsiness and each methodology with its own limitations.

II. RELATED WORKS

A. Related Works

Fouzia et al. [7] proposed a system that presents a drowsiness detection structure subject to shape marker computation, that perceives the eyes, and besides checks the eye gleam rate followed by drowsiness detection at nonstop. In the proposed structure, the bits of knowledge in regards to the eye status is traversed picture handling computations, which offer a non-meddlesome method for managing perceive drowsiness with no bother and block. In future, the detection of yawning of the driver can be moreover be done using same packaging work for distinguishing further experiences with respect to the drowsiness of driver. Exactly, the proposed framework is to identify shut eyes to notice drivers exhaustion and caution the driver with a ringer and vibration on sure detection.  During observing, the framework can conclude whether the eyes are open, shut or tired. At the point when the eyes were recognized shut for a really long time, an admonition signal is given. This was finished by mounting a camera before the driver and persistently catching its continuous video involving Open CV in Raspberry Pi. Driver drowsiness not set in stone from a few side effects that manifest in tired drivers face. Through the investigation of the eye status, the framework will actually want to tell regardless of whether the driver is tired. At first when the camera is in on state, video transfers are consistently caught from the drivers face. To distinguish the eye squint, the present status of the eye is required which is either open or shut.

Assuming the condition of eyes changes from shut to open, it demonstrates an eye squinting. In the event that the condition of the eyes is in shut state for a specific measure of time then the individual is recognized as sleepy and shape indicator is utilized to anticipate the condition of the driver's eye. In the event that a tired driver is recognized a signal sound and the vibration is raised, until the driver is ready. To assess the proposed strategy, Raspberry pi-B module 3 is utilized as a preprocessor, which processes the pictures.

Raspberry pi has a SD card embedded into the space on the board which goes about as the hard drive for the Raspberry Pi. It is controlled by USB and the video result can be connected to a customary RCA TV set, a cutting edge screen, or even a TV utilizing the HDMI port. The web camera is utilized to distinguish the eye of the driver, the information gathered from the camera is shipped off the Raspberry pi, after information handling, on the off chance that the driver is tired, he/she will the alarmed with vibrator which is set under the driving seat.

Federico Guede-Fernández et.al. [8] proposed a method to give a sureness quality level of the respiratory sign. Likewise, the obtained quality sign level has been gotten together with the drowsiness detection computation to further develop the course of action results through diminishing the amount of fake up-sides in light of changes of assessed RRV related not to drowsiness yet rather body advancements or talking. Besides, the arranged computation has been endorsed under a driving test framework and a couple of drowsiness scenes have been recognized for each and every one of 15 test gatherings. The best quality degree of drowsiness events has been delivered by external onlookers from video annals of the subject while driving. The show results have been gotten following LOSOCV framework to achieve a fair-minded check of the summarized computation execution. The improvement of drowsiness course of action results due to the sign quality computation has been moreover studied and the qualifications for each headway models and AT have been discussed.

To overview the precision of the proposed drowsiness ID estimation, the workers of the examination ought to drive while they are engaging against to fall asleep. Since these conditions may be unsafe for volunteers, the test was vehicle ried on a driving test framework hold up under controlled conditions. Also, test conditions have been expected to see the lead of drivers in their fight against to fall asleep while driving. From one perspective, the preliminary show is based on drowsiness disclosure while driving in striking hard circumstances to keep status: night hours, debilitating and somewhat less jam-stuffed streets. This show has been set up with a front screen inside a test framework with vehicle bodywork to give a distinctive contribution with the workplaces of the Organization of Biomechanics (IBV) in Valencia, Spain as shown in Fig. 1a. The vehicle was outfitted with pedals, coordinating deal transmission.

A projector was used to show what is happening on a screen before the vehicle. The preliminaries were guided with the room climate control to 24?C, low lighting and with turnpike sounds. The reenactment circumstance was a two-course street with two ways toward each way, low thickness of traffic, night environment, and way with no sharp curves. Moreover, a camcorder, which has been revolved around the subject's head, has recorded the assessment. The camera device used to record video was the Logitech webcam C120. Video accounts were used by a couple of outside observers to deliver the ground truth driver drowsiness signal. Individuals have played out the driving tests in two unmistakable days. In one of them, the subject had rest difficulty for the last night so subjects have not snoozed in the 24 hours before the examination. The other examination happened with the customary rest situation, so subjects likely refreshed somewhere near 6 hours the night preceding the test. The two circumstances were randomized and the tests were performed at the hour of day at 9 a.m. Then, the differentiation between the two examinations was the subjects' fundamental rest condition. Subjects were drawn closer to remain in the test framework vehicle seat, wearing a safety belt and keeping two hands on the coordinating wheel with the two feet in the pedals. During the underlying 5 minutes of the test, subjects were drawn closer to remain still and quiet. Starting there ahead, they were drawn closer to drive the test framework vehicle for an hour and a half. Finally, they were drawn nearer to remain arranged in the driving test framework with their eyes shut for 5 minutes.

Akshay Bhaskar et al. [9] proposed EyeAwake in its stream stage makes a fair device for slow driver detection and giving fundamental vehicle change similar to moving down and over the long haul stopping the vehicle. The model cost generally US$40 to develop and test. To keep the cost low, an essential arrangement was followed which uses principal yet exact sensors. By using different such sensor sections, detection of slow driving was truly exact. Starting on road tests show a 70% precision in distinctive sluggish driving. Furthermore, EyeAwake has wonderful features that most existing things could do without outside notice to various drivers and individuals by walking, and a central vehicle update instrument. Right away, all of the sensors were autonomously pursued for needed convenience. Heaps of change was finished in this stage to kill anything number false up-sides as would be reasonable. Note that this arrangement is done with respect to particular drivers and can't be summarized.

This is a direct result of the way that different drivers have different characteristics, for instance, complexion, height, resting heart/breathing rate, etc Contingent upon the driver's physiology and lead, the breaking point regards referred to in the above portion are set. After all sensors completed the above appraisal, they were composed to shape EyeAwake. The EyeAwake model was first attempted in a lab under ideal working circumstances. On an ordinary, 80% accuracy was achieved by EyeAwake in the lab. In any case, several counterfeit up-sides sorted out some way to slither into the structure. After lab testing, EyeAwake was conveyed in vehicles which were driven by carriers and cabbies during night shifts. The truck and cabbies were at first incredulous as they had been significant for similar thing tests previously. Regardless, with an accuracy of around 70% on street, they were interested by the thing. The two basic features of EyeAwake they appreciated were: I. the ability to ease off and in the end stop the vehicle ii. outside admonitions to various drivers and individuals by walking while sluggish driving was perceived. Additionally, they gave positive reviews concerning the accommodation of EyeAwake and the insignificant cost. Xiaoxi Mama et al. [10] proposed a driver weakness detection system subject to CNN using significance video groupings is proposed. To involve the spatial and common information for significance based movement affirmation issue, another designing called significance video-based two-move CNN is proposed. Likewise, an establishment removal structure for significance video plan of driving is proposed.

The proposed methodology can feasibly recognize whether the driver is exhaustion driving or customary driving during the night, with a precision of 91.57%, using our assembled significance based driver weariness dataset. Our structure can be used to give alerts suitably to weariness drivers, which will decrease and thwart the occasion of vehicle crashes. Aldila Riztiane et al. [11] proposed a structure that wanted to caution drivers so they can be encouraged to pull over and quit driving in a lethargic state. The application "Driver Drowsiness Detection" utilizes Haar-course Detection similarly as format planning in OpenCV to perceive and follow the eyes using the front camera of an Android contraption. Testing has been coordinated to ensure that the convenience, lead, execution and client satisfaction are exactly as expected. Regardless of the way that, the data got by the application really has a couple of restrictions, expressly in relationship to satisfactory lighting and cloudiness of the face and eyes an area, the application has successfully distinguished the eye glimmers at the place of 30 to 60 degrees and distance of 20-50 cm, similarly as assessing the beat. The system plan for drowsiness revelation application can be summarized as follow.

In any case, the driver will be drawn closer to measure their heartbeat using the wearable device to get the normal resting beat, as it contrasts for every individual. At the point when the assessment has been saved, it's everything except a limit to conclude the value showed as would be normal, Tired, or Sleeping. These characteristics will be secured after the beat assessment testing to conclude the drop regard from common heartbeat to tired.

At the point when the assessment is taken and saved, the driver will really need to start the application which will perceive the face, and basically the eyes space of the driver. The application will distinguish the state of the eyes, whether or not it is shut or open. The two limits, which are the beat and the state of the eyes will then be united to choose if the driver is arranged to being lethargic or not. To run the utilization of Driver Drowsiness Location, there are two devices required.

The first is an Android-based PDA with front camera, and second an Android Wear-based wearable contraption. There are moreover endorsed gear and programming subtleties to ensure that the application runs as cared about. The contraptions used to run Driver Drowsiness Application ought to be organized not solely to ensure the most vital efficiency, yet moreover not disturbing the client, which requirements to use the application while driving. The principal contraption, which has the front camera, is placed in front with the objective that it's everything except an undeniable viewpoint on the client's facial features.

The distance and mark of the circumstance will be settled following testing is driven. The device can be placed some put on the dashboard of the vehicle using a phone holder unequivocally expected to be used in the vehicle. The design organizing with methodology returns a value called minVal to find the most dark space of an image, which on the off chance that there ought to be an event of a face, is the student of the eyes. Thus, the way toward recognizing squint could rely on the presence of such worth. The value will not be returned when the eyes are closed, since it's everything except distinguished. Exactly when the eyes are open, it will go from 0 to 5.

The minVal will then be one of the closing circumstances to set off the wariness. Regardless, the squint range also ought to be considered to choose Drowsiness. The glimmer range of a sluggish individual should be around 0.6 seconds, in this way the value that will be used for the application will be 0.6 seconds. A beginning clock limit will be started to count whether the minVal has exited go after a period of 0.6 seconds. Right when such condition has been met, the application will check for another condition, which is the beat, to choose Drowsiness and setting off the application. Since there is no exact heartbeat regard that can be applied as a standard to every individual, testing for the sole justification for this application will be coordinated to choose the drop of an individual's heartbeat between three stages; prepared, sluggish, and dozing. The beat regard, assessed in beats every second (bpm) will be one more focal thought to set off the caution.

Sanya Gupta et al. [12] proposed a work which is attempted and checked to give high precision in the dynamic of the driver's lethargic circumstances. All the while, the system accomplished a couple of execution limits of continuous plans in vehicles. The system gave in this paper is helpful, since it needn't bother with a power source from a PC and ought to be related with a PC right at the hour of arrangement of the structure. Since the structure manages the norm of equivalent handling which certainly diminishes the handling time required by the system to give the decision, it makes it a continuous course of action. Additionally, FPGA is a versatile contraption that can be absolutely reconfigured for higher or better computation. The working of individual modules was attempted freely similarly as all together, and the results were gotten exactly as expected.

The incline twist gives the minima on x position connecting with one side of the face and maxima on x position contrasting with the right 50% of the face.

The fact that the eye is open makes the eye acknowledgment module yield is transported off Intricacy module where the multifaceted nature of the image is evaluated, differentiated and the edge and a decision given. For head cutting down module, the chief framework is removed and ready. Rest of the housings are moreover arranged similarly and number of face pixels for every circumstance are diverged from the essential edge with conclude head cutting down. One such taken care of edge with head cut down. The system made in this paper is attempted and checked to give high accuracy in the dynamic of the driver's apathetic circumstances.

All the while, the system accomplished a couple of execution limits of nonstop plans in cars. The structure gave in this paper is helpful, since it needn't bother with a power source from a PC and ought to be related with a PC right at the hour of arrangement of the system. Since the system works on the norm of equivalent getting ready which most certainly reduces the dealing with time required by the structure to give the decision, it's everything except a steady plan. Furthermore, FPGA is a versatile contraption that can be absolutely reconfigured for higher or better estimation. Notwithstanding, the test that remains is to make the structure sagacious using smaller than normal controllers since FPGA is an extravagant contraption. Furthermore, the system made anticipates that the groundwork of the driver should be essential.

If a bewildered groundwork is presented behind the driver, the structure will in everyday make a misstep in the decision. One more drawback of the structure is that it doesn't consider head moving to be a sign of driver drowsiness yet it gives obvious results to the extent that disturbing the redirected driver, turning his head left or ideal for term past the breaking point time. So the future work will contain killing this heap of drawbacks and troubles and taking other sign of drowsiness like yawning and head moving into thought too. Rajamohana S.P. et al. [13] proposed a system which is based on hybrid approach of convolutional neural network and bidirectional long short term memory (CNN_BILSTM). The proposed work is to perceive facial milestones from pictures are caught as the individual is driving the vehicle, through a camera mounted onto the vehicle and give the data procured to the certified model to distinguish the state of the driver. In the event that the information gathered is distinguished to give indications of drowsiness of the driver would be advised to stop the vehicle to forestall mishaps.

The crossover approach of CNN BiLSTM is proposed for open eye detection of driver drowsiness. The exhibition of the proposed technique is satisfactory and the utilization of a web camera during the evening time could be gotten to the next level. The way to deal with CNN BiLSTM has been displayed to work actually in low-goal pictures for the overwhelming majority eye stances. The information assortment was parted into 70% for arrangement and 30% for testing, covering all phases of eye openings and eye-glass tests. Data about the drivers head and eye position are gotten through shape indicator with Euclidean calculations. This technique yields phenomenal outcomes on pictures with changing foundations and lighting. The proposed framework can hence identify whether your eyes are opened or shut. In the event that eyes are shut for a really long time, a telling call might be shipped off the driver. This framework identifies sleepiness and cautions the driver by telling a bring continuously.

Conclusion

The frameworks which have been proposed till currently are not particularly dependable on the grounds that those frameworks are not effective to extricate the valuable data. The data is feeling the loss of the delicate edges that trail the exactness to accomplish the right acknowledgment. The vast majority of the frameworks depend on single edge choice and not investigating the subsequent casings for right choice. Framework required appropriate investigation for achieving the best degree of precision. Programmed Drivers Drowsiness Detection is a helpful idea for executing safe driving that may mindful drivers for an unplanned setbacks. The point of this study is to address an answer for one of the significant reasons for the street mishap, the driver drowsiness; the proposed arrangement tracks the driver\\\'s eyes and afterward tell him when his eyes set shut everything down try not to lose the control of the vehicle and causing traffic mishaps. On the off chance that the consequence of the characterization demonstrates that the driver\\\'s eyes is shut for a predefined timeframe, the eyes of the driver will be viewed as shut and consequently a caution will be begun to alarm the driver. Exactness is vital as for the right and mistaken acknowledgment for ideal framework. In future, a framework can be executed that relates the best exactness with negligible blunder rate that works successfully at continuous. This framework can be improved in future where exactness depends by affixing different tensorflow based pressing that works successfully in PC vision particularly continuously.

References

[1] U.S. Department of Transportation, “Intelligent Vehicle Initiative 2002 annual report,” http://ntl.bts.gov/lib/23000/23500/23572/13821.pdf [2] National Highway Traffic Safety Administration. Research on Drowsy Driving. Accessed October 20, 2015. [3] W. Wierwille, L. Tijerina, S. Kiger, T. Rockwell, E. Lauber, and A. Bittne, “Final report Supplement – heavy vehicle driver workload Tech, Rep, No.Dot Hs 808 467(4) (1996) [4] Dehnavi, M. ; Attarzadeh, N. ; Eshghi, M., \\\"Real time eye state recognition\\\" IEEE May 2011 [5] Jaeik Jo Sung Joo Lee, Ho Gi Jung, Ryoung Park, Jaihie Kim \\\" vision based method for detecting driver drowsiness and distraction monitoring system\\\" Optical Engineering Vol 50(12) December 2011 [6] Linkedin, Driver Drowsiness Detection Alert System with Open-CV & Keras Using IP-webCam For Camera Connection, Available: Sep 2020, https://www.linkedin.com/pulse/driver-drowsiness-detection-alert-system-open-cv-keras-khandave/. Accessed: 29 June 2021. [7] Fouzia, R. Roopalakshmi, J. A. Rathod, A. S. Shetty and K. Supriya, \\\"Driver Drowsiness Detection System Based on Visual Features,\\\" 2018 Second International Conference on Inventive Communication and Computational Technologies (ICICCT), 2018, pp. 1344-1347, doi: 10.1109/ICICCT.2018.8473203. [8] F. Guede-Fernández, M. Fernández-Chimeno, J. Ramos-Castro and M. A. García-González, \\\"Driver Drowsiness Detection Based on Respiratory Signal Analysis,\\\" in IEEE Access, vol. 7, pp. 81826-81838, 2019, doi: 10.1109/ACCESS.2019.2924481. [9] A. Bhaskar, \\\"EyeAwake: A cost effective drowsy driver alert and vehicle correction system,\\\" 2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS), Coimbatore, India, 2017, pp. 1-6, doi: 10.1109/ICIIECS.2017.8276114. [10] X. Ma, L. Chau and K. Yap, \\\"Depth video-based two-stream convolutional neural networks for driver fatigue detection,\\\" 2017 International Conference on Orange Technologies (ICOT), Singapore, 2017, pp. 155-158, doi: 10.1109/ICOT.2017.8336111. [11] A. Riztiane, D. H. Hareva, D. Stefani and S. Lukas, \\\"Driver Drowsiness Detection Using Visual Information On Android Device,\\\" 2017 International Conference on Soft Computing, Intelligent System and Information Technology (ICSIIT), Denpasar, Indonesia, 2017, pp. 283-287, doi: 10.1109/ICSIIT.2017.20. [12] S. Gupta, P. Jain and E. Rufus, \\\"Drowsy Driver Alerting System,\\\" 2018 Second International Conference on Electronics, Communication and Aerospace Technology (ICECA), Coimbatore, India, 2018, pp. 1665-1670, doi: 10.1109/ICECA.2018.8474931. [13] S.P. Rajamohana, E.G. Radhika, S. Priya, S. Sangeetha, [14] Driver drowsiness detection system using hybrid approach of convolutional neural network and bidirectional long short term memory (CNN_BILSTM), Materials Today: Proceedings, Volume 45, Part 2, 2021, Pages 2897-2901, [15] F. You, X. Li, Y. Gong, H. Wang and H. Li, \\\"A Real-time Driving Drowsiness Detection Algorithm With Individual Differences Consideration,\\\" in IEEE Access, vol. 7, pp. 179396-179408, 2019, doi: 10.1109/ACCESS.2019.2958667. [16] B. M. K. Kumari and P. R. Kumar, \\\"A survey on drowsy driver detection system,\\\" 2017 International Conference on Big Data Analytics and Computational Intelligence (ICBDAC), Chirala, Andhra Pradesh, India, 2017, pp. 272-279, doi: 10.1109/ICBDACI.2017.8070847. [17] A. Pinto, M. Bhasi, D. Bhalekar, P. Hegde and S. G. Koolagudi, \\\"A Deep Learning Approach to Detect Drowsy Drivers in Real Time,\\\" 2019 IEEE 16th India Council International Conference (INDICON), Rajkot, India, 2019, pp. 1-4, doi: 10.1109/INDICON47234.2019.9030305. [18] M. Miranda, A. Villanueva, M. J. Buo, R. Merabite, S. P. Perez and J. M. Rodriguez, \\\"Portable Prevention and Monitoring of Driver’s Drowsiness Focuses to Eyelid Movement Using Internet of Things,\\\" 2018 IEEE 10th International Conference on Humanoid, Nanotechnology, Information Technology,Communication and Control, Environment and Management (HNICEM), Baguio City, Philippines, 2018, pp. 1-5, doi: 10.1109/HNICEM.2018.8666334. [19] Z. Jie, M. Mahmoud, Q. Stafford-Fraser, P. Robinson, E. Dias and L. Skrypchuk, \\\"Analysis of Yawning Behaviour in Spontaneous Expressions of Drowsy Drivers,\\\" 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018), Xi\\\'an, China, 2018, pp. 571-576, doi: 10.1109/FG.2018.00091. [20] J. Yu, S. Park, S. Lee and M. Jeon, \\\"Driver Drowsiness Detection Using Condition-Adaptive Representation Learning Framework,\\\" in IEEE Transactions on Intelligent Transportation Systems, vol. 20, no. 11, pp. 4206-4218, Nov. 2019, doi: 10.1109/TITS.2018.2883823. [21] S. Kusuma, J. Divya Udayan and A. Sachdeva, \\\"Driver Distraction Detection using Deep Learning and Computer Vision,\\\" 2019 2nd International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT), Kannur, India, 2019, pp. 289-292, doi: 10.1109/ICICICT46008.2019.8993260. [22] Y. Wang, L. Jin, K. Li, B. Guo, Y. Zheng and J. Shi, \\\"Drowsy Driving Detection Based on Fused Data and Information Granulation,\\\" in IEEE Access, vol. 7, pp. 183739-183750, 2019, doi: 10.1109/ACCESS.2019.2960157. [23] C. Yang, X. Wang and S. Mao, \\\"Unsupervised Drowsy Driving Detection With RFID,\\\" in IEEE Transactions on Vehicular Technology, vol. 69, no. 8, pp. 8151-8163, Aug. 2020, doi: 10.1109/TVT.2020.2995835. [24] M. M. Islam, I. Kowsar, M. S. Zaman, M. F. Rahman Sakib and N. Saquib, \\\"An Algorithmic Approach to Driver Drowsiness Detection for Ensuring Safety in an Autonomous Car,\\\" 2020 IEEE Region 10 Symposium (TENSYMP), Dhaka, Bangladesh, 2020, pp. 328-333, doi: 10.1109/TENSYMP50017.2020.9230766.

Copyright

Copyright © 2022 Hemant Ahirwar, Minal Saxena. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.