A Web-based Attendance System Using Face Recognition

2.1 Literature Review

Facerecognitionremainsthemostpopularamongvarious means of identification, such as fingerprints [1], RFID [2], passcodes,andsoon.Facerecognitionisstilldevelopingto predict[3]andverifytheperson,evenwhentheothersare accurate.AccordingtothepresentCOVID 19scenario,itis thediscussedareSectionandinteraction.monitoringgiveFurthermore,facialcreatedorderputsattendaessentialtohaveabasiccontactsystemandkeepmeticulousncerecords.Theoldmethodofcollectingattendanceusersindangeroftransferringdiseasesandviruses.Intofindasolution,thiswebbasedapplicationwastoofferacontactlessattendancesystemthatusesarecognitionmodeltotrackstudents'attendance[4].itautomatesallavailablestatisticalanalysestoinstructorsandstudentshasslefreeattendanceandeliminatestheneedforanypersonalInpart2,thestudyopenswithanintroductionaliteraturereview.Thesuggestedsystemisdescribedin3.Insection4,thesuggestedsolutionandapproachpresentedindetail.TheexperimentalfindingsareinSection5.Intheconcludingportion,section6,studyconcludeswithaconclusionandfuturescope.

2. RELATED WORK

p ISSN: 2395 0072

e ISSN: 2395 0056

International Research Journal of Engineering and Technology (IRJET)

Volume: 09 Issue: 03 | Mar 2022 www.irjet.net

Thesystemmustbebuiltusingareal timefacialrecognition modelthatofferscorrectservicewiththefewestpotential differences.Afterexaminingthedifferentmodelsstatedin [5][6][7][9][13][14], it was determined that the 'python3

Mr. Rajvardhan Shendge1, Mr. Aditya Patil2, Mrs. Tejashree Shendge3

Abstract Covid 19 has generated a paradigm change in day to dayworkthathasneverbeenseenbefore.Intoday's efficienttheattendance,userrecommendations.Furthermore,techniqueuploadingmanuallytechniquesmustrequired.environment,socialdistanceandlimitedphysicalcontactareDespitethesechallenges,educationalinstitutionscontinuetooperate.Mostinstitutions'existingfordocumentingclassroomattendanceentailputtingthestudent'ssignatureonpapersandthenthemintothesystemforfurtheranalysis.Thisiscurrentlyoutdatedandantiquated.itisharmfulsinceitcontradictsthecontactlessThisarticleoffersamethodinwhichamayutilizefacialrecognitiontoregisterhisorherandoursystemwillautomaticallygenerateallofdatafortheinstructorstoevaluateinapainlessyetway.

1, 2Student, Computer Engineering, Ramrao Aidik Institute of Technology(India)

Key Words: AttendanceSystem,FacialRecognition,Web basedapplication,Onlineclassroom,Opensource

A Web-based Attendance System Using Face Recognition

1.INTRODUCTION

face recognitionprogrambuiltbyAdamGeitgey[4]offered students,institutionprovideswithattendance.student'staughtmanualthis'GooglecomputeradditionaldexactingBeThereTocompatibility.preliminaryareas,ideaandfasterAninsufficient.analyticsaboveThisFaceresultingdatabasemodelhistogramsAalerts.tosystem'sForattendancereliablesuggestedfacialartandFacesthegreatestaccuracy.ThereportedaccuracyontheLabeledintheWildbenchmarkdatasetis99.38percent[4],thispackageisconstructedutilizingdlib'sstateofthefacerecognition,adeeplearningbasedarchitecture.Forrecognition,[5]usesa20layeredCNNmodel.Theapplication[5]fallsshortwhenspecifyingaserveranddatabasearchitectureforrealtimemanagement.facedetection,[6]usestheDlibCNNlibrary.Thisdataanalyticsareinaccessible,andthesystemfailscommunicatewithitsstakeholdersthroughemailsorhaarcascadeclassifierpairedwithlocalbinarypattern(LBPH)wasusedtocreatethefacerecognitionin[7].ItisaSQLdatabasethat'sbeenutilized.Thearchitectureoftheproposedsystem[7]ispoor,inrestricteddataanalysesrecognizersHaarCascadeandLBPHareutilizedin[9.studycontinuesworkingonimprovingtheaccuracyto90%andmakingthegadgetmoreportable.Thedataofferedtothecompany'sstakeholderswereenhancedAlexnetmodelisutilizedin[14],whichhasatrainingtime(109ms),fewernetworkparameters,afasterfeatureextractiontime(98ms).Althoughtheisintriguing,theprogramasawholefallsshortinmanyincludingthelackofauserfriendlyinterface,dataanalysis,andlackofmultiplatformaddresstheinadequaciesofthepreviousexamples,presentsacomprehensiveapplicationwithandatabasemanagementsystem.Itprovidesindepthataanalyticstoitsconsumerstopresentthemwithinformation.Italsosupportsmobileanddevicesonamultidevicebasis.Furthermore,aClassroom'[8]likearchitectureisimplementedintosystemtorelievestudents'relianceandinstructorsfromattendancemarkingforawidevarietyofcoursesinvariousbatches.ThisstructureallowseachenrolmenttobemarkedindependentlyforTheclassroomarchitecture,whencombinedafacerecognitionsystemforrecordingattendance,aplethoraofopportunitiesfortheeducationaltokeepacalculativeandstatisticalrecordofteachers,attendance,absentees,massbunks,

3Student, Electronics and Telecommunication Engineering, Fr. C. Rodrigues Institute of Technology(India) ***

Most commercial face analysis methods favor one race, ethnicity, culture, age, or gender over another [10]. The systemusesAdamGeitgey'spythonfacialrecognitionmodel [4].ThemodelisbasedonLabeledFacesInTheWild(LFW) andhasa99.38percentaccuracyrate.TheLFWdatasethas a uniform distribution of demographics [10], including a widevarietyoffacesbelongingtodiverseethnicities,races, and genders, which helps to reduce bias in visual face Manyidentification.onlineapplicationsarestillconcernedaboutcross site reference forgery(CSRF)[11].Thesuggested systemuses CSURF,anexpress.jsmiddleware,tolimitthepotentialofa data breach resulting in the dissemination of sensitive information, particularly user photos. Each user request produces a random CSRF token and stores it as a session cookie for that user's authentication. Furthermore, all sensitivedataissafeguardedbystoringsensitivedataasa addresshash,suchaspasswordsorpersonalinformationlikeahomeandphonenumber[12].



Recognition layer/ML layer: This layer employs a pre trained model to extract and recognize facial characteristics from the current real time video frame given by the camera module and record attendanceinthedatabase.



The training layer entails training the face recognition model using training photos for face identification,extractingthemodel'sfeatures,and storingtheminadatabase.



Volume: 09 Issue: 03 | Mar 2022 www.irjet.net



p ISSN: 2395 0072

3. PROPOSED SYSTEM

Joiningaclassroom:Studentsmayjoinaclassroom by entering a unique code given to them by their subjectinstructor.

onlinecoursework,andemailfacilitythatexistingsystems system.[5][6][7][9][13][14]failtoprovidecollectivelyinonesingleFig1.LiteratureSurvey

3.1 SYSTEM ARCHITECTURE

Fig.2.SystemArchitecture

BeThereconsiderstwomainstakeholders:Teacherandthe students.ForTeachers,BeTherehasfollowingkeyfeatures:



Weblayer:Savetheindicatedattendanceina databasesothatAPIsmaybeusedforfuture analysis.

Filtering and Downloading: Teachers may sort students by Yea, Batch, and Class, as well as downloadthesystem'scolor codedattendanceand absentee,excelsheets.



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

Monthly analysis: A graphical depiction of the supplied.attended,numberoflecturesgiven,thenumberofpeoplewhoandthenumberofmassbunkedpeople



3.2 SCOPE

Defaulterslist:Studentsmayseetheattendanceof courseshighlightedinredbecausetheirattendance islessthantherequiredlevelof75.

2.2 Mitigating bias in visual recognition and data security

Creating a one of a kind code for student enrollmentintheclassroom:Eachtopicinstructor may generate a one of a kind classroom code for their subject and share it with their students for essentialsharingcoursework.BeThereincludesthefollowingcharacteristicsforstudents: classesmayCheckaverageattendanceforallcourses:Astudentviewhisorheraverageattendanceforallinwhichheorsheisenrolled.

Attendance recording: Teachers can accurately record each student's attendance as well as the defaulters.aclass'saverageattendance.Teachershaveaccesstolistofallenrolledpupils,aswellasalistof



faces.attendanceinthedatabasebasedonthoserecorded



Based on a poll as part of requirement elicitation, the important characteristics listed in Table II were implemented. Five instructors who took part in the poll

instructorbybothStatisticsandgraphs:Themonthlygraphanalysisoftotallecturesperformedandtotalmassbunksstudentsareshownonboththestudentandportals.

Fig.3.TechnologyStack

User Registration/Authentication: When a user accessestheportalforthefirsttime,heorshemust register.Allinputfields,suchastheemailfield,the roll number field, and the password field, are verified during registration. Students and instructors register individually to keep track of usecrucialinformationforfuturereference.Usersmustthesamecredentialseverytimetheylogin.

4.DatatabseERDiagram

5. EXPERIMENTAL RESULTS



4.2 System Modules



Face Recognition[4]: After a student has been ofherenrolled,thismoduleisinchargeofcollectinghisorattendance.Itopensacameraandtakesthefacethestudentinthatframe,thensavesthestudent's

Handlebar is a templating engine that allows frontendandbackendAPIs tocommunicatewithdynamic data.Express.jsisaNode.jsbackendframeworkthatisused to create APIs. A python flask server hosts the facial recognitionmodel[4].Opencvisalsoincorporatedintothe Python server to record real time camera feeds. All of the dataiskeptintheMongoDBdatabase,whichallowsseveral userstoviewandwritethesamedatasimultaneously.

p ISSN: 2395 0072

4. PROPOSED SYSTEM AND METHODODLOGY



Attendance module: Once a class is formed, this module calculatestheaverageattendance of each student in each enrolled class, the average attendance of all courses, and separates the defaulters'list.

Themainentitiesinthedatabaseareusers,classes,images andattendance. Theinternal relationshipoftheseentities canbedetailed

ThetechnologiesusedtocreateBeThereareshowninthe diagram below (Fig2). A javascript server (nodejs and expressjs), a python flask server, and a NoSQL MongoDB source.environmentV8databasearethemaincomponentsofthetechnology.Ontheengine,Node.jsisabackendJavaScriptruntimethatiscrossplatformandopen

4.1 TECHNOLOGY STACK



The diagram below (Fig3) represents the database Entity Relationshipforthissystem.TheDatabaseisaNoSQLbased databasei.eMongodb(Fig2).

Classrooms: Teachers may establish topic classrooms using a unique class code and then manuallyor,usingthecode, addstudents to each class.

Emailnotification:Notifiesstudentswhentheyare addedtoanewclassbyateacherorwhentheyare placed in the defaulters' category (attendance below75)

Fig.as,

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

Volume: 09 Issue: 03 | Mar 2022 www.irjet.net

4.3 DATABASE DESIGN

Fig.8.Classroomspageforteachers

p

Fig.7.DashboardforTeachers

Volume: 09 Issue: 03 | Mar 2022 www.irjet.net ISSN: 2395 0072

Fig.6.Multiplefacedetectioninanonlineclassroom

This web based application (BeThere!) offers a variety of extensivefeaturesthathaveyettobeexploredinsystems like this[5][6][7][9][13][14]. A comprehensive automated experience is provided by the smooth integration of face recognition and custom APIs and the interaction between studentsandinstructorsthroughthewebapplication.Itis also a goal to continue improving this application in the future. Because this is an open source project, the open sourcecommunitywillgivemoredevelopmentiterationsto improvetheapplication'susabilityandusefulness.Various preventingexperiments,suchasUIchanges,modularadvancementsforspoofingwhenregisteringattendance,and

provided information on data analysis and features they would want to see in this system. Table III shows the through11RAaccuracysystemofmeetingCOVIDaftersystem'smajorperformancemetrics,whichwerecomputedextensivetestingwithaclassof20pupils.Becauseof19rules,theexperimentwasdoneusinganonlineprogram.Themeetingprogramallowedforatotalninestudentstobeseeninoneview(Fig.4),andthecorrectlyrecognizedthestudentswithanaverageof98percent.A2.3GHztwocoreCPUwith8GBMwasusedtotestthesystem.Figures4,5,6,7,8,9,10andshowtheoutcomesofateacherpresentingalectureanonlinemeetingapplicationusingawebbrowser.Fig.5.FEATUREANALYSISandPERFORMANCEANALYSIS

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

6. CONCLUSION AND FUTURE SCOPE

time defects.developingAnyonehasstandardsepidemic,providingspoofdetectingincreaselowerasadditionalbeing.Futureworkwillusetheclassroomschematoincludeinstructorandstudentinteractionselements,suchquizzesandfilesharing.Experimentingandtestingtothelatencyofdifferenttaskslistedbeforemaythesystem'sperformance.Severalmodelsforlivelinesswillbeevaluatedtomakethesystemproof.Theapplicationwascreatedinthehopesofasafe,educationalatmosphereinthiscoronavirusaccordingtotheregulatoryauthorities'safetyandrequirements.ProjectURL:ThetechnologybeenimplementedopensourcebyTeamBeThere!maydownloadandinstallthesystem,contributetonewfeaturesbyfilingapullrequest,andreportFig.9.StudentDashboardFig.10.Individualsubjectclassroomwithstatistics

REFERENCES

Volume: 09 Issue: 03 | Mar 2022 www.irjet.net ISSN: 2395 0072 have been deferred for the

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

[2] Murizah Kassim, Hasbullah Mazlan, Norliza Zaini, Muhammad Khidhir Salleh, “Web based Student AttendanceSystemusingRFIDTechnology”,2012IEEE Control and System Graduate Research Colloquium (ICSGRC2012),doi:10.1109/ICSGRC.2012.6287164

[1] MohamedHammad1,YashuLiu1andKuanquanWang1, “Multimodal Biometric Authentication Systems Using ConvolutionNeuralNetworksbasedonDifferentLevel FusionofECGandFingerprint”,IEEEAccess(Volume:7), doi:10.1109/ACCESS.2018.2886573.

Fig.13.Absenteelistforaparticularday

p

Fig.11.Defaulterstudentslistpageforteachers

Fig.12.Exportedattendancexlsxsheetforteachers;color codingschemaused (Bad(lessthan50%):Red, Average(50 60%):Orange,Good(60 75%):Yellow)

[3] IacopoMasi,YueWu,TalHassner,andPremNatarajan, “DeepFaceRecognition:aSurvey”,201831stSIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI),doi:10.1109/SIBGRAPI.2018.00067.

[10] AshrafKhalil,SohaGlalAhmed,AsadMasoodKhattak, Nabeel AlQirim, “Investigating Bias in Facial Analysis Systems:ASystematicReview”,IEEEAccess(Volume: 8),2020.doi:10.1109/ACCESS.2020.3006051.

[8] IzwanNizalMohdShaharanee,JastiniMohdJamil,and Sarah Syamimi Mohamad Rodzi, ”The Application of Google Classroom as a Tool for Teaching and Learning“,Vol.8No.10,e ISSN:2289 8131.

[9] HarikrishnanJ,AryaSudarsan,AravindSadashiv,Remya Ajai A S, “Vision Face Recognition Attendance MonitoringSystemforSurveillanceusingDeepLearning TechnologyandComputerVision”,2019International Conference on Vision Towards Emerging Trends in Communication and Networking (ViTECoN), doi: 10.1109/ViTECoN.2019.8899418.

[14] Wenxian Zeng, Qinglin Meng, Ran Li, “Design of IntelligentClassroomAttendanceSystemBasedonFace Recognition”, 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), doi: 10.1109/ITNEC.2019.8729496.

Volume: 09 Issue: 03 | Mar 2022 www.irjet.net p ISSN: 2395 0072

[5] SyamKakarla,PriyaranjanGangula,M.SaiRahul,C.Sai CharanSingh,andT.HitendraSarma,“SmartAttendance ManagementSystemBasedonFaceRecognitionUsing CNN”, 2020 IEEE HYDCON, doi: 10.1109/HYDCON48903.2020.9242847.[

[7] Yuslinda Wati Mohamad Yusof, Muhammad Asyraf MohdNasir,KamaAzuraOthman,SaifulIzwanSuliman, Shahrani Shahbudin, Roslina Mohamad, “Real Time Internet Based Attendance Using Face Recognition System”, International Journal of Engineering Technology, 7(3.15)(2018)174 178, doi: http://dx.doi.org/10.14419/ijet.v7i3.15.17524

[11] W.H.Rankothge,SM.N.Randeniya,“Identificationand MitigationToolforCross SiteRequestForgery(CSRF)”, 2020 IEEE 8th R10 Humanitarian Technology Conference (R10 HTC), doi: 10.1109/R10 HTC49770.2020.9357029.

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

[12] Friedrich Wiemer, Ralf Zimmermann, “High speed implementation of bcrypt password search using special purpose hardware”, 2014 International ConferenceonReConFigurableComputing andFPGAs (ReConFig14),doi:10.1109/ReConFig.2014.7032529.

[6] TataSutabri,Pamungkur,AdeKurniawan,andRaymond Erz Saragih, “Automatic Attendance System for University Student Using Face Recognition Based on Deep Learning”, International Journal of Machine Learning and Computing, Vol. 9, No. 5, October 2019, doi:10.18178/IJMLC.

[13] Dongmei Feng, Peng Wang, Lei Zu, “Design of Attendance Checking Management System for College

ClassroomStudentsBasedonFingerprintRecognition”, doi:2020ChineseControlAndDecisionConference(CCDC),10.1109/CCDC49329.2020.9164638.

[4] Adam Geitgey, “Machine learning is fun!: The world’s easiestintroductiontomachinelearning”,AdamGeitgey Medium2014, https://medium.com/@ageitgey/machinelearning is fun 80ea3ec3c471,Lastretrievedon8thAugust,2021.

Turn static files into dynamic content formats.

Create a flipbook