Gyro sensor based smart helmet for automated early accident detection

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

Gyro sensor based smart helmet for automated early accident detection

Ayan Bin Rafaih

Aitchison College, Lahore, Pakistan

***

Abstract

The number of worldwide motorcycle accidents is increasing every year, which necessitates the need of safety procedures such as wearing helmets. The integration of Internet of Things (IoT), Artificial Intelligence (AI), Automated Detection and Machine Learning (ML) into this field has improved safety and efficiency, with the help of smart helmets. Our approach proposes an Arduino-based helmet that provides early accident-detection through the use of a gyro sensor to measure the accelerations in all directions. This system continuously monitors for any variations from the gyro offsets and the upper and lower thresholds defined for each direction (x, y and z): a positive detection would lead to the GSM module connecting to a pre-defined SIM network, along with the GPS module for location coordinates, to send a message to the user’s emergency contact as soon as the accident is detected. Thus, the message would act as an early warning system that would also include the map latitude and longitude coordinates. This setup provides high accuracy and precision, with constant monitoring of the angular velocities at a set frequency.

Key Words: Smarthelmet,Accidentdetection,Gyrosensor,ArduinoGSM,ArduinoGPS

1.INTRODUCTION

Rapidurbanizationgloballyhasledtoasignificantincreaseinthenumberofmotorvehicleaccidentsthatoccureveryyear, especiallymotorcycleaccidents.Thesesevereaccidentsoftenleadtounusuallyhighmortalityratesandmorbidityresultingin temporaryorpermanentdisabilities.Forexample,28,356motorcycleaccidentswerereportedovera6-yearperiodbythe LegalMedicineOrganization[1]whichincludedsevereandnon-severefatalitiessuchasheadtrauma,fracturesandinternal bleeds.SimilarstatisticscollectedfromstudiesinKenyarevealed1073motorcycleaccidentsovera6-monthperiodthatledto patientsunderintensivecareatthehospitals,withamorbidityrateof6.8%[2].Themostfrequentandmajorinjurytothe people admitted to the hospitals after motorcycle accidents is head injury, with the lack of the helmet being the primary contributingfactortothehighmortalityrate.Incomparison,motorcyclistswearinghelmetsarelessfrequentlyinvolvedinlifethreateningheadinjurysituationsasthehelmetsprovideanactiveprotectivemechanism.

Mostofthepeopleusingamotorcycleindevelopingandunderdevelopedcountriesdon’twearahelmet,whichputsthemat risk for severe head injury in the event of an accident. A lack of awareness of traffic rules and road limits, inadequate infrastructuraldevelopmentsandnon-adherencetospeedlimitsleadtofatalinjuriesamongmotorcyclists.Overspeedingleads tohigherimpactvelocity,increasingthechancesofhead,chest,abdominalandlimbinjuries.Inmanyofthesecountries,there havebeenamultitudeofsafetystrategiesandpoliciesthathavebeen introducedyethavenotbeenfullyimplemented.As coveredintheWorldHealthOrganizationGlobalStatusReport[3],trafficdeathshavecontinuedtorisesincethepastdecade, whichdirectlyobstructsthegoalofreducingtrafficaccidentsandinjuriesby2030.Roadtrafficinjuryhasremainedaleading causeforthedeathofyoungpeopleaged30orless[3].

Helmetstypicallyconsistofanoutershellmadeofathermallystablematerialtoprotectfromcombustionfire.Thisislinedon theinsidebyanenergy-absorbinglayerofmaterialtohelpreducetheimpulsiveforceexperiencedontheheadduetoasudden fall.Followingthatisacomfortlayerconsistingofadditionalpolymersalongwithastraptohelpprovidefirmgriptothehead [4]. Motorcyclists wearing helmets are particularly less likely to have severe head injuries that may lead to permanent disabilities.Therehavebeenmultipleapproachestoaddressthisissue.Forexample,governmentssuchasinBangladesh[5] have taken stringent actions such as increasing awareness and legislations regarding driving limits for motorcyclists. Additionally,therehavebeeneffortstointegratetheInternetofThings(IoT)intothisfieldbymakingsmarthelmetsthatwould helpindetectingaccidentsuchasdescribedbyShabbeeretalandKumaretal[6][7].However,someofthesehelmetsprovide false detection of accidents or the trigger mechanism for the micro-controllers don’t work as expected - accidents are sometimesnotdetectedproperlyandtimelyforthemotorcyclists.Intheeventofheadinjuryfrommotorcycleaccident,timeto treatmentbytheemergencyservicesiscritical.Anearlydetectionandresponsebytheemergencyservicesleadtobetter clinicaloutcomesforthepatientthusreducingthemorbidityandmortality. Ifimplementedproperly,smarthelmetscanhelp reducethenumberofaccidentsthathappenannuallybyagreatamount andsavemanylivesthroughthisprocess.Smart

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

helmets,ifcombinedwithdeeplearningtechniques,helptoprovidemanyusefulfeaturessuchasmessaging,accidentdetection andreal-timelearning.

OurapproachinthisstudyutilizesArduinocomponents,anautomatedsatellitemessagingsystemandacontinuousangular velocitysensortotakereadingsthatwouldbeprocessedbythemicro-controller.Thesevalueshelpdeterminewhenthehelmet wearingmotorcyclisthasfallen off– specificaccidentpositionreferenceframesaredefinedthatwouldberead andthen processedtodetermineiftherehasbeenanaccident.Themessagingsystemhelpstoautomaticallynotifytheuser’semergency contactoftheoccurrencesothattheycanbegivenimmediatemedicalattention.OurproposedhelmetutilizesaGlobalSystem forMobileCommunication(GSM)moduleandGlobalPositioningSystem(GPS)moduletohelpsendthemessagecontainingthe accurateGPScoordinatestoidentifythelocationoftheaccident.Basedonpre-definedtiltsandrotationsduringanaccidentin eachreferenceframeforthehelmetandapre-settimer,ifthehelmetsatisfiesthecriteriasetforaccidentdetection,thenthe messagewouldautomaticallybesent.Thisstudypresentsanew,simplifiedapproachtomakesmartearlyaccident-detection helmetsusingArduinocomponents.

2. LITERATURE REVIEW

Therehavebeenmanyeffortstohelpproposedifferentideasforsmarthelmetsthatcouldbeutilizedforaccidentprevention. Someofthosefocusonusingforce-detectionsensorswhileothersuseaMachineLearning(ML)approach.Forexample,Rasli [8]presentedanapproachwherethemicroprocessorprocessedthedatacollectedbytheForce-SensingResistance(FSR)anda speed sensor (BLDC Fan), along with a radio frequency module for signal transmission between the circuits used. The mechanismsetupensuredthattheengineofthemotorcyclewouldnotstartuntilthehelmetwasfullywornandthespeed sensorwouldlimitthemaximumsafespeedforthemotorcyclist.Similarly,Chandranetal[9]proposedasmarthelmetthat usedanaccelerometersensor,measuringtheaccelerationoftheheadcontinuouslyinallcartesiandirectionsx,yandz.A conditionalcheckwasenabledonthevalueswhichifexceedthesetvaluewouldcorrespondwiththeWIFIEnabledmodule thatwouldbeabletocommunicateasenseofemergencywhichalsoincludestheuseofcloudcomputinginfrastructures.

Furthermore,therehavebeeneffortsbyRahman[5]tocombinetheInfraredsensortodetectthepresenceofthehelmetonthe head,alongwithanalcoholsensortodetectthelevelofalcoholinthedriver’breadth:thiswasconnectedtoaGraphicalUser Interface(GUI)applicationandalertsystem.A reviewbyImpanaetal[10]foundthatmostsmarthelmetsmostlycomprisedof twomainparts;amotorcyclepartandahelmetpartwheretheformerwasoftenconnectedtothebatterycircuitoftheengine preventingitfromturningitonincaseofanyabnormalitiesandthelatterconsistedofanarrayofsensors,suchasHall-Effect Sensors(HES)forpositioningandspeeddetectionusingtheprincipleofthechangeinvoltageduetoitspresenceinamagnetic field.However,suchapproachesheavilyrelyonthepresenceofapowersupply,whichindeedwasnotideal. Mhatreetal[11] utilizedasimilarmodularapproachfortheirdesignofasmartaccidentdetectionsystem.Theirbikemoduleusedavibration sensor to take readings on a constant sampling rate to be sent to the micro-processor through the receiver-transmitter communication channel. Kurkute [12] used a unique approach. Instead of using Arduino or a PIC controller, they use a RaspberryPiModule,apressuresensorandimageprocessingalgorithms.TheyusedtheHaarCascadesystemwithfrontalface tracking,contourtrackingalgorithmsandcannyedgedetectionalgorithmwithgaussianfilterstocarryoutimageprocessing. Themethodusedaverificationsystemwhosefilterswouldoutputwhetherintheimagethedriverwaswearingthehelmetor not,byfocusingonlyonrelevantpixeldataandshades.Inconjunctionwiththepreviousapproach[5],astudybyAhuja[13] used a tilt and a NC sensor to create a sensitive infrared system, which however faced problems with tilting as the microcontrollerwasnotfullyabletoprocesswhenthetiltisgreaterthantheboundariesset.

Jeong[14]usedamixtureofmultipledetectionandrelaysystemstohelpincreasetheaccuracyoftheaccident-detection.They didthisbyusinganelectro-opticalcameraforthesamepurposedefinedbyKurkute[12],oxygenresidualsensorstodetectthe driver’shealthinthecaseofanemergencyandalterthecommandcentre,a smart watchthataidsinthecommunication processanda6-axisinertialsensortodetecttherelativemotionofthedriver’shead.Behr[15]proposedthehazarddetection elementofthehelmetwornbydrivers,especiallyminers,inathree-foldmanner;preventionagainsthelmetremovalusingan Infrared(IR)Sensortodetectwhenit’sontheheadandwhenit’snot,collisiondetectionincasetheminersarestruckorbitby anobjectcausinganinjurythatexceedsthevalueof1000intheHeadInjuryCriteria(HIC)anddetectingtheconcentrationof harmfulgasesintheatmospherebytheuseofmultiplegassensorsandthefinalcommunicationbetweenthecomponentsand emergencyalertissentusingZigBee,insteadofWIFIorBluetoothinthepaper. Desai[16]usedafalldetectionalgorithm governedbyOpticalCharacterRecognitionSystem(OCRS)thatwouldsendanautomaticalerttothenearesthospitaland emergencyservices,intandemwithahelmetdetectionalgorithmgovernedbyaHoughTransformDescriptor(HTD).Apart fromthis,therehavealsobeensomeapplicationsofMachineLearning(ML)andDeeplearning(DL)suchastwo-dimensional andthree-dimensionalconvolutionalneuralnetworks(CNNs)toperformmulti-tasklearningasdescribedintheapproachby Linetal[17].Theycreatedanewlocaldatasetusingtheincidentsintheirareas,whichwasthenusedforsimilaritylearningin eachincidentandhelmetuseclassification,tobesubsequentlyevaluatedthroughmetricssuchasprecision,accuracyand

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

ReceiverOperatingCharacteristic(ROC)Curves.Similarly,Vishnuetal[18],usedCNNsfordetectionofmotorcycledriverswho werenotwearinghelmetbyfeatureextractionfromthepixelsandheadlocalization.

3. METHODOLOGY

OurAccident-detectionhelmetusesaGyrosensortohelpdetecttheoccurrenceofaccidents.Thefollowingsection exploresthetechnicalapparatususedinourhelmet.

3.1. MPU 6050 Gyroscope and Accelerometer Sensor

TheMPU6050sensormoduleconsistsofa3-axisgyroscopesensorcomponentwithMicroElectroMechanicalSystem(MEMS) technologyandisusedtodetectandflagrotationaltiltsineitherofthecartesianaxesbymeasuringthevalueoftherotational velocity. For example, in our case when the motorcycle driver’s head tilts through an accident and the helmet lands in a disturbedpositionratherthanlandingonitsbase,thegyroscopedetectsarotationaroundthex,yandzaxes.Thisrotationis causedbytheCorioliseffectwhichproducesadetectablevibrationfortheMEMSinsidethemodule.WecandefinetheCoriolis force(Fc)thatleadstogyrosensor’sworkingas: Fc =-2m( xv’)

Wheremisthemassofthebody(inourcasethehelmet),  istheangularvelocityofthehelmetandv’isthetangentialvelocity ofthehelmetrelativetotheinertialreferenceframe.Thisissupplementedbythe3-axisaccelerometerintheMPU6050which helpstodetecttheangleoftiltthathasoccurredrelativetothenormalframeandinitialcoordinatepositions.Angulartilts displacethemovablemassfromitsoriginalposition,whichisregisteredas anoutputamplitudefromthecapacitorinthe module.Thesetwomoduleshelptocalculatethedegreeoftilt  anddirectionoftilt.Thus,wecanobtaintheangularvelocity andtherelativeaccelerationalongeachoftheaxes,whicharecontinuouslysampledbythemicroprocessor.

3.2. Arduino Nanoboard

WeusetheArduinoprovidedNanoboard,whichhastheATmega328Pmicrocontrollerfromthe8-bitAVRfamily,whichutilizes an operating voltage of 5V for our helmet and has a 2KB Static Random Access Memory (SRAM). The micro-controller is directlyinterfacedandconnectedtoboththesensor(thereceiverendforallthedatavalues)andothercomponentstoprovide complete functionality. The sensor data is compared with the upper and lower limits set to detect when an accident has occurred,andthisisdonecontinuously,withitspowersupplyconnectedtoarechargeablebattery.

3.3. Global System for Mobile Communication (GSM) Module

WeuseanautomatedShortMessageService(SMS)systeminthecaseofanaccidentafterthedatafromthegyroscopesensoris analyzed. For this purpose, we utilize a SIM800L GSM Module. Operating on a quad-band GSM network, it connects and communicateswiththeinternetusinga TransmissionControl Protocol/InternetProtocol (TCP/IP).Theuserwouldhave providedtheprogramwithanemergencynumberinthecaseofanaccident.TheGSMmodulewouldautomaticallysendaSMS messageonthatnumber,lettingtheemergencycontactknowthattheuserhasbeeninvolvedinanaccident.Theinformationof themessagewouldcontaintwoelements:anindicationthattherehasbeenanaccidentandthemapcoordinateswhichwould beprovidedbytheGPSinterface.GSM’santennaisconnectedtoitsNETpin,whiletheArduinomicrocontroller’sTransmitPin (TXD)isconnectedtoitsReceiverPin(RXD).TheTXDpindefinedis11andRXDis10.

3.4. Global Positioning System (GPS) Module

Tosupplementthemessagethatissenttotheemergencycontact,thelocationofwheretheaccidenthasoccurredisalsosentin themessage.Forthis,weusedtheNEO-6MGPSmodulewithantennathatwouldregisterandsendtwocartesiancoordinates: longitudeandlatitude.Thesecoordinateswouldbesentalongwiththewarningsystem,whichcanbeprocessedbyanymap apptoshowthelocation.

3.5. LM2596 DC-DC Step Down Power Supply Module

WeusetheLM2596Buck-ConvertormoduletohelpstepthebatteryvoltagedowntoasuitablevoltagefortheSIM800LGSM Modulesoitsvoltageisinitsrequiredrange.Thevoltageissteppeddownfrom8voltstobetween3.3and4.2volts.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

3.6. Buzzer System

An Arduino piezo buzzer is also attached which would make a beep sound when the message is sent and an accident is successfullydetectedbythemicrocontroller.

4. WORKING PRINCIPLE

TheArduinoNanoboardandmoduleisconnectedtoarechargeablebatterywhichactsasitspowersource.TheMPU6050 moduledeterminesthegyrooffsetstomaketheinertialreferenceframe,whicharecrucialfordeterminingtherelativeangular velocitiesinthenextpart.AcheckisintroducedthatensuresthatGx,GyandGzarecalculated,thegyrooffsetsforthex,yandz directionsrespectively.Ifthegyrooffsetsarenotcalculated,itwillrecalculateandinitializethereferenceframeafteradelayof 500milliseconds.Thegyrosensormeasurestheangularvelocityinallthecartesiandirectionsrelativetoanormalinertial referenceframeforthehelmetandthemotorcycledriver’shead.TheseangularvelocitiesaredifferentiatedtocalculateAccX, AccY andAccZ,theaccelerationsforthex,yandzdirectionsrespectively.

Boundarythresholdvaluesareintroducedtodeterminewhentheaccelerationineachdirectiongooutofbounds.Theupper andlowerboundsaredeterminedthroughbrute-forceexperimentation–Anytiltof  greaterthan15degreesisclassifiedtobe avalidcasetodeterminethevaluesforthegyroaccelerations.Thesegyroaccelerationsarecollectedandthen analyzedto determinethebasevalue.Maximumandminimumfluctuationseithersideofthemiddle,assuminganormaldistribution,are noted and help form the set of possible accelerations that should lead to the correct classification of an accident thus eliminatingtheoccurrenceoffalsepositivesignals.Forbothaccelerationinthex-directionandinthez-direction,ifAccXorAccZ isgreaterthan0.40orlessthan-0.40,orforaccelerationinthey-direction,ifAccYisgreaterthan0.45orlessthan-0.45,aflag wouldberaised.Acountvariable,thathasalreadybeeninitializedimmediatelyaftertheflag,wouldincrementsequentiallyper second.When30secondshaveelapsed,theprocedure“EmergencyMessage ( )”wouldbecalled.Abuzzerwouldalsosoundin thehelmettoalertanypeoplewalkingnearby.IfthevaluesofthevariablesAccX,AccY andAccZ arenotoutofthethreshold values,thentheloopwoulditerateindefinitely iftheArduinoNanoboardisturnedonthroughtheattachedswitchinthe helmet.

TheProcedure“EmergencyMessage ( )”consistsofamainprocedureaswellasanin-builtprocedure.Thein-builtprocedure wouldbelinkedtotheGPSModule–connectingwiththesatellite.TheGPSmodulewouldintroduce2globalvariablesintothis procedure: Longitude and Latitude measures which correspond to the location of the accident on a map. For the main procedure,theSIMnetworkwouldbecontactedthroughtheuseoftheGSMmodule.Theemergencynumberwouldalreadybe savedintheprocedureandfinallythemessageissenttothespecifiedcontactnumberwithawarningpromptsuchas“Accident Detected”. Theworkingprincipleisexplainedin figure 1 and figure 2.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

5. RESULTS

Aseriesoftrialswereconductedtohelpsetaccuratethresholdvaluesfortheaccelerationsderivedfromthegyrosensor.These preliminarytrialshelpeduschoosethethresholdvaluesforeachaccelerationvalue (table 1)

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

Table 1: Thresholdvaluesforaccelerationinx,yandzdirections

Withtheconditionalstatementdefinedinthemethodologysection,thishelmetactsasasuccessfulearlyaccidentdetection method.TheNEO-6MGPSModuleusedinthisstudyhasatrackingsensitivityof-162dBm,whichhelpsinbettertracking,ifthe accidentsoccur.TheMPU6050Gyrosensorusedforangularvelocitiesandaccelerationshasaveryhighaccuracy-thevalues aresampledeachsecondandcontinuouslycheckedfortheaccidentcondition.Thesensorhasabandwidthrateof100Hz,with itsrespectivetoleranceband.Oncetheaccidentisdetected,thelocalnetworkfortheprovidedphonenumberiscontacted through which a message is generated as a flag for the occurrence of the accident. This message is sent through the SIM networkwiththelocationcoordinates.

6. CONCLUSION

Theresultsandmethodologyofthisprojectpresentanew,simplifiedapproachtomakingasmartaccident-detectionhelmet thatcanbeaccessibletoeveryone.Thispaperhaspresentedanovelwaytomakeahelmetthroughbasicmaterials,whichwill help reduce the amount of severe injury people often receive when they are driving a motorcycle, especially in remote locations.Ourhelmetenablesimmediatemedicalcareandalertstofamilymembersforallmotorcyclists.Additionally,theuse oftheGPSmoduleshelpstoidentifytheexactlocationoftheaccident.

Theprojectcanbeadaptedwithreinforcementlearning,adaptivefeedbackordeeplearningtechniqueswithback-propagation toenablelive-detectionofaccidentsandlive-learningofnewsituations.Suchimprovementscouldincreasethecaseswhereit flagsanaccidentcorrectlyandimproveitsaccuracyreducingthenumberoffalsepositivesignalsgenerated.

7. REFERENCES

1. AbdolrazaghBarzegar,MasoudGhadipasha,MehdiForouzesh,SamiraValiyari,AliKhademi,Epidemiologicstudyof trafficcrashmortalityamongmotorcycleusersinIran(2011–2017),ChineseJournalofTraumatology,Volume23,Issue 4,2020,Pages219-223.

2. Wilberforce Cholo, Wilson Odero, and Japheths Ogendi. "Epidemiological characteristics of crashes and pattern of motorcycle injuries presenting to hospitals in Kisumu City, Kenya: a cross-sectional study." Pan African medical journal 45.1(2023).

3. World Health Organisation “Global Status Report on Road Safety 2023” https://www.who.int/teams/socialdeterminants-of-health/safety-and-mobility/global-status-report-on-road-safety-2023

4. MichelleAiello,UgoGalvanetto,andL.Iannucci."Numericalsimulationsofmotorcyclehelmetimpacttests." International journal of crashworthiness 12.1(2007):1-7.Rahman,MdAtiqur,etal."Iotbasedsmarthelmetandaccidentidentification system." 2020 IEEE Region 10 Symposium (TENSYMP).IEEE,2020.

5. MdAtiqurRahman,S.M.Ahsanuzzaman,IshmanRahman,ToufiqAhmed,andAbidAhsan"Iotbasedsmarthelmetand accidentidentificationsystem." 2020 IEEE Region 10 Symposium (TENSYMP).IEEE,2020.

6. Shoeb Ahmed Shabbeer, and Merin Meleet. "Smart helmet for accident detection and notification." 2017 2nd International Conference on Computational Systems and Information Technology for Sustainable Solution (CSITSS).IEEE, 2017.

7. M.KiranKumar,AniruddhaBalbudhe,andCHSaiKarthikeya."SmartHelmetbasedAccidentDetectionandNotification SystemforTwo-WheelerMotorCycles." E3S Web of Conferences.Vol.391.EDPSciences,2023.

8. Mohd Khairul Afiq Mohd Rasli, Nina Korlina Madzhi, and Juliana Johari. "Smart helmet with sensors for accident prevention." 2013 international conference on electrical, electronics and system engineering (ICEESE).IEEE,2013.

9. SreenithyChandran,SnehaChandrasekar,andN.EdnaElizabeth."Konnect:AnInternetofThings(IoT)basedsmart helmetforaccidentdetectionandnotification." 2016 IEEE Annual India Conference (INDICON).IEEE,2016.

10. H. C. Impana, M. Hamsaveni, and H. T. Chethana. "A review on smart helmet for accident detection using IOT." EAI Endorsed Transactions on Internet of Things 5.20(2019).

11. KimayaMhatre,RajNandwadekar,AdityaPatil,RushikeshShinde,andPradnyaKamble"SmarthelmetWithIntercom Feature." Proceedings of the 3rd International Conference on Advances in Science & Technology (ICAST).2020.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 12 Issue: 01 | Jan 2025 www.irjet.net p-ISSN: 2395-0072

12. Swapnil Kurkute, Nikita Ahirao, R. G. Ankad, and V. B. Khatal "IOT based smart system for the Helmet detection." Proceedings of International Conference on Sustainable Computing in Science, Technology and Management (SUSCOM), Amity University Rajasthan, Jaipur-India.2019.

13. Prashant Ahuja, Ketan Bhavsar. "Microcontroller Based Smart Helmet Using GSM & GPRS." 2018 2nd International Conference on Trends in Electronics and Informatics (ICOEI).IEEE,2018.

14. Mingi Jeong, Hyesun Lee, Myungnam Bae, Dong-Beom Shin, Sun-Hwa Lim, and Kang Bok Lee "Development and application of the smart helmet for disaster and safety." 2018 International conference on information and communication technology convergence (ICTC).IEEE,2018.

15. C.J.Behr,AnujKumar,andGerhardP.Hancke."Asmarthelmetforairqualityandhazardouseventdetectionforthe miningindustry." 2016 IEEE International Conference on Industrial Technology (ICIT).IEEE,2016.

16. Maharsh Desai, Shubham Khandelwal, Lokneesh Singh, and Shilpa Gite."Automatic helmet detection on public roads." International Journal of Engineering Trends and Technology (IJETT) 35.5(2016).

17. HanheLin,JeremiahD.Deng,DeikeAlbers,andFelixWilhelmSiebert."Helmetusedetectionoftrackedmotorcycles usingcnn-basedmulti-tasklearning." IEEE access 8(2020):162073-162084.

18. ChalavadiVishnu,DineshSingh,C.KrishnaMohan,andSobhanBabu."Detectionofmotorcyclistswithouthelmetin videosusingconvolutionalneuralnetwork." 2017 International joint conference on neural networks (IJCNN).IEEE,2017.