MULTIPLE FACE DETECTION SYSTEM USING DEEP LEARNING

AbstractThe smart classroom leverages automation to streamline tasks such as attendance registration, whichtraditionally require considerable time and effort. Conventional methods—including identificationcards, radio frequency systems, and biometric technologies—often face limitations related to safety,accuracy, and cost. However, recent advancements in digital image processing, particularly facerecognition technology, present a promising alternative.This study introduces an automated attendancesystem utilizing the YOLOv8 algorithm, capable of detecting and recognizing multiple student facessimultaneously with high efficiency. The system was tested on a real time dataset and achieved up to90-95% accuracy, highlighting its reliability and effectiveness in automating attendance processes.The proposed system not only automates the attendance marking process but also generates analyticalreports. Face recognition plays a vital role in uniquely identifying individuals, making it an idealsolution for classroom attendance automation through the integration of advanced face detectiontechniques. Keywords: Attendance, Face Recognition, HOG, LBPH, MySQL, YOLOv8, CNN