Face Recognition System using OpenCV

Abstract

Face Recognition is an important computer vision technology that enables automatic identification ofindividuals based on facial attributes. It suffers from problems like variations and facial movements. This project utilizes OpenCV-based techniques for real-time detection and recognition of faces from real video streams. The system combines Haar Cascade for face recognition and Local Binary Histogram (LBPH) for identification to offer absolute identification. The system also automates the retrieval of student academic records by correlating identified faces to the database, offering efficiency and security in student identification. Combining face recognition with web automation simplifies the process of result retrieval, minimizing human error and processing time. This project solves shared weaknesses of conventional biometric authentication systems in a contactless, scalable, and secure manner. The system is well-suited to school applications in attendance monitoring and student record keeping,but is generally applicable to other uses ofauthentication.

Introduction

Face recognition technology has revolutionized biometric authentication by providing a contactless, efficient method for identity verification, addressing the limitations of traditional student verification systems prone to inefficiencies and fraud. Core technologies like the Haar Cascade Classifier for face detection and the Local Binary Pattern Histogram (LBPH) for feature extraction underpin modern systems, enhanced further by deep learning models such as DeepFace, VGGNet, and FaceNet, which have significantly improved accuracy and scalability.

Despite advancements, challenges like pose variation, lighting conditions, and security threats remain. Researchers have proposed improved face alignment, multi-task learning, and liveness detection to boost reliability and security. Face recognition is widely used in surveillance, access control, and educational institutions for attendance tracking and student verification, often integrated with AI and cloud systems for scalable, automated record management.

The project described implements a real-time face recognition system using Haar Cascade and LBPH combined with Selenium-based web automation to authenticate students and automatically retrieve academic records. It employs hardware like Intel Core i5 processors and cameras (720p+), and software tools such as Python, OpenCV, and SQLite/MySQL for database management. The system achieves high accuracy (96.3% in normal light), fast processing (~200ms), and low error rates, ensuring reliable, secure, and scalable student authentication with real-time attendance tracking and academic data access.

Conclusion

Computer vision, oneofthesubsets ofartificialintelligence,allowsimagestobe processed andmeaningfulfeaturesextracted by acomputer.TheapplicationofHaarCascadewasextremelyefficient,evenincaseswheresubjectsworeglasses,forfacedetection, and LBPH offered an inexpensive yet consistent face recognition method. The support for real-time processing ensured smooth system operation with no perceptible frame delay. The OpenCV Face Recognition System combines face recognition and web automation, as well as foregoing physical student identification. By utilizing OpenCV for recognition and detection, the system provides an effective and fast way of verifying students in real time. Once identified successfully, the system will automatically fetch academic records from the university portal with the help of Selenium webscrapingwith minimalhuman intervention. Important detailslikeCGPA,hallticketnumber,andcoursesenrolled are fetched with very little user intervention. The use ofcontemporarytechnologies like Python, OpenCV, Selenium, andSQLite/MySQL makesthesystem scalable,secure, and efficient. Thesolutionimprovesthemanagementofacademicaffairsthroughauto-academicrecordretrieval,minimumhumanintervention, and maximum precision. Lastly, the project solves thetask of manual verification of students andrecords, yielding a reliable and state-of-the-art solution toeducationalinstitutions.Itssuccessthrough combiningcomputer vision with webautomation meansthatother solutions using such features,e.g., universityidentityverification ina moreautomaticform,securitysystems,andcompanyoffices, can alsobe contemplated.

References

[1] P.ViolaandM.Jones,\"RapidObjectDetectionusingaBoostedCascadeofSimpleFeatures,\"IEEEConferenceonComputer Vision and Pattern Recognition (CVPR), 2001. [2] T. Ahonen,A. Hadid,andM. Pietikainen,\"FaceRecognition with Local BinaryPatterns,\" European Conferenceon Computer Vision (ECCV), 2004. [3] Y.Taigman,M.Yang,M. A.Ranzato,andL.Wolf, \"DeepFace: ClosingtheGaptoHuman-Level Performance,\"Facebook AI Research, 2014. [4] O.M.Parkhi,A.Vedaldi,andA.Zisserman,\"DeepFaceRecognition,\"OxfordUniversity,2015. [5] F.Schroff,D.Kalenichenko,andJ.Philbin,\"FaceNet:AUnifiedEmbeddingforFaceRecognitionandClustering,\"Google AI Research, 2015. [6] Z.Zhang,Z.Zhang,andX. Liu,\"JointFaceDetection andAlignment,\"IEEEConferenceonComputerVisionandPattern Recognition (CVPR), 2016. [7] A.AlotaibiandM.H.Mahoor,\"AgeandGenderClassificationusingFaceRecognition,\"IEEE,2017. [8] R. Ranjan, V. M. Patel, and R. Chellappa, \"Multi-Task Learning forFace Recognition,\"IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. [9] XLiandX.Li,\"3DFaceRecognitioninSecurityApplications,\"IEEE,2017. [10] E.Lopez,P.Ruiz,andM.García,\"FaceRecognitioninSurveillanceSystems,\"IEEE,2018. [11] A.BalabanandR.Huber,\"LivenessDetectioninFaceRecognition,\"Springer,2018. [12] P. BoykoandA.Sukharev, \"PerformanceEvaluationandComparisonofOpenCVandDlibfor FaceRecognition,\"IEEE, 2019. [13] C.WuandQ.Liu,\"FacialExpressionAnalysisforRecognitionImprovement,\"Elsevier,2019 [14] S.SrivastavaandP.Rathi,\"SmartAttendanceSystemUsingOpenCV,\"InternationalJournalofEngineeringResearch & Technology (IJERT), 2020. [15] J.KempandM.Richardson,\"FacialRecognitioninEducationSystems,\"ScienceDirect,2020. [16] S.Sharma,\"FaceRecognitionforSecureAuthentication,\"Elsevier,2021. [17] A.Pateletal.,\"AutomatedStudentIdentityVerificationandAcademicRecordRetrievalUsingAI,\"IEEEAccess,2023. [18] XZhaoetal.,\"Cloud-BasedFaceRecognitionfor ScalableStudent Authentication Systems,\"ACMTransactionson Cloud Computing, 2023.

Copyright

Copyright © 2025 T Venkata Subbamma, Jaswanth Reddy Asodi, Peesari Harshitha, Sai Kumar Rao, Gona Saicharan. 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.