Implementasi Face Recognition pada Sistem Presensi Mahasiswa Menggunakan Metode Single Shot Multibox Detector dan Local Binary Pattern Histogram

MUHAMMAD, FAUZAN YASYKUR (2024) Implementasi Face Recognition pada Sistem Presensi Mahasiswa Menggunakan Metode Single Shot Multibox Detector dan Local Binary Pattern Histogram. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (2MB)
	Text Abstract.pdf Download (197kB)
	Text Abstrak.pdf Download (204kB)
	Text BAB I.pdf Download (220kB)
	Text BAB II.pdf Download (696kB)
	Text BAB III.pdf Download (1MB)
	Text BAB IV.pdf Restricted to Registered users only Download (1MB)
	Text BAB V.pdf Download (204kB)
	Text Daftar Pustaka.pdf Download (215kB)
	Text Lampiran.pdf Restricted to Registered users only Download (663kB)

Abstract

The digitalization era has profoundly impacted diverse sectors, including higher education. Continuous technological innovation is crucial to supporting academic activities, notably the student attendance system. Implemented technologies, like RFID on Student ID Cards and QR code scanning, still bear weaknesses such as the risk of ID card loss, QR code misuse, and proxy attendance. Biometric systems, particularly face recognition, emerge as alternative innovations to enhance the authenticity of student attendance. Studies on face recognition, using methods like SSD, Fisherface, Eigenface, and LBPH, reveal SSD's excellence in face detection but lesser effectiveness in face recognition compared to LBPH, which performs reasonably well. This research addresses fraud prevention in the attendance system and enhances face recognition accuracy by implementing real-time technology in a web-based student attendance system. The combination of Single Shot Multibox Detector (SSD) and Local Binary Pattern Histogram (LBPH) methods achieves this. SSD functions as the face detector, and LBPH as the face recognizer. The study evaluates accuracy in detecting and recognizing student faces based on parameters like distance, the number of faces in a frame, and face position. Face detection tests within a distance radius of 30 cm to 100 cm and various face positions result in 100% accuracy. Face recognition tests based on face position achieve an 84,8% accuracy and 81,5% precision, while those based on distance parameters yield an 85% accuracy and 82,1% precision. Keywords: Face Recognition, Attendance System, SSD, LBPH

Actions (login required)

View Item