RANCANG BANGUN SISTEM PRESENSI BERBASIS PENGENALAN WAJAH MENGGUNAKAN ANDROID

ARKAN AFADHLAN, . (2026) RANCANG BANGUN SISTEM PRESENSI BERBASIS PENGENALAN WAJAH MENGGUNAKAN ANDROID. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (553kB)
	Text BAB 1.pdf Download (226kB)
	Text BAB 2.pdf Restricted to Registered users only Download (639kB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (634kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (468kB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (214kB) | Request a copy
	Text DAFTAR PUSAKA.pdf Download (212kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (811kB) | Request a copy

Abstract

Sistem presensi manual pada perguruan tinggi rentan terhadap manipulasi data, tidak efisien, dan membutuhkan waktu pencatatan yang lama. Untuk mengatasi kendala tersebut sekaligus mendukung efisiensi administrasi smart campus, dikembangkan aplikasi UNJ Electrical Student Attendance System berbasis Android. Menggunakan bahasa Python serta pustaka OpenCV dan DeepFace, sistem pengenalan wajah (face recognition) ini dirancang untuk memverifikasi identitas mahasiswa secara otomatis, cepat, dan presisi. Proses perancangannya mengaplikasikan metode Research and Development (R&D) berbasis pendekatan Forward Engineering serta model SDLC Waterfall yang meliputi tahap analisis, perancangan, koding, pengujian, dan pemeliharaan. Aplikasi ini mengintegrasikan fitur deteksi wajah, pemantauan lokasi, pemrosesan real-time, dan rekapitulasi laporan. Kehadiran fitur-fitur tersebut mempermudah mahasiswa melakukan absensi secara mandiri serta memfasilitasi dosen dalam mengelola administrasi kelas. Evaluasi fungsional (black box testing) mengonfirmasi bahwa teknologi ini beroperasi secara optimal pada kondisi cahaya yang memadai dan efektif mencegah kecurangan titip absen. Tingkat kelayakan aplikasi juga terbukti sangat tinggi, ditunjukkan oleh pencapaian skor 98% dari validator ahli media dan 94% dari responden mahasiswa. Hasil ini menegaskan bahwa sistem presensi berbasis pengenalan wajah ini sangat layak diimplementasikan sebagai media pendukung pembelajaran digital yang jauh lebih unggul dibandingkan sistem konvensional. ***** Manual attendance systems in higher education are susceptible to data manipulation, inefficient, and time-consuming. To address these limitations while supporting the administrative efficiency of a smart campus, the Android-based UNJ Electrical Student Attendance System application was developed. Utilizing Python alongside the OpenCV and DeepFace libraries, this face recognition system is designed to verify student identities automatically, quickly, and precisely. The design process applies the Research and Development (R&D) methodology based on a Forward Engineering approach and the Waterfall SDLC model, which encompasses analysis, design, coding, testing, and maintenance phases. The application integrates face detection, location tracking, real-time processing, and report recapitulation features. The presence of these features facilitates students in performing self-attendance and assists lecturers in managing classroom administration. Functional evaluation (black box testing) confirms that the system operates optimally under adequate lighting conditions and effectively prevents proxy attendance fraud. The application's feasibility rate is also proven to be exceptionally high, demonstrated by a score of 98% from expert media validators and 94% from student respondents. These results confirm that this face recognitionbased attendance system is highly feasible for implementation as a digital learning support medium, significantly outperforming conventional systems.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr. Aris Sunawar, M.T. ; 2). Dr. Imam Arif Rahardjo, M.T.
Subjects:	Teknologi dan Ilmu Terapan > Teknik Energi > Energi Listrik
Divisions:	FT > S1 Pendidikan Teknik Elektro
Depositing User:	Arkan Fadhlan .
Date Deposited:	31 Aug 2026 04:02
Last Modified:	31 Aug 2026 04:02
URI:	http://repository.unj.ac.id/id/eprint/73509

Actions (login required)

View Item