Adliansyah, Fadel Najmi (2025) SISTEM ABSENSI BERBASIS FACE RECOGNITION DENGAN MODEL INCEPTION-RESNET. S1 thesis, FAKULTAS TEKNIK UNIVERSITAS SULTAN AGENG TIRTAYASA.
![[img]](https://eprints.untirta.ac.id/style/images/fileicons/text.png) |
Text (Fulltext)
Fadel Najmi Adliansyah_3337210008_Fulltext.pdf Restricted to Registered users only Download (1MB) |
|
Text (Bab 1)
Fadel Najmi Adliansyah_3337210008_01.pdf Restricted to Registered users only Download (584kB) |
|
Text (Bab 2)
Fadel Najmi Adliansyah_3337210008_2.pdf Restricted to Registered users only Download (270kB) |
|
Text (Bab 3)
Fadel Najmi Adliansyah_3337210008_03.pdf Restricted to Registered users only Download (255kB) |
|
Text (Bab 4)
Fadel Najmi Adliansyah_3337210008_04.pdf Restricted to Registered users only Download (1MB) |
|
Text (Bab 5)
Fadel Najmi Adliansyah_3337210008_05.pdf Restricted to Registered users only Download (189kB) |
|
Text (Daftar Pustaka)
Fadel Najmi Adliansyah_3337210008_Ref.pdf Restricted to Registered users only Download (186kB) |
|
Text (Lampiran)
Fadel Najmi Adliansyah_3337210008_Lamp.pdf Restricted to Registered users only Download (316kB) |
|
Text (Cek Plagiasi)
Fadel Najmi Adliansyah_3337210008_CP.pdf Restricted to Registered users only Download (13MB) |
Abstract
This study aims to develop an automated attendance system based on facial recognition. The system was developed using the Waterfall methodology. The system leverages the Inception-ResNet architecture and is trained using the VGGFace2 dataset to accurately recognize user identities. The training process was conducted over 8 epochs, resulting in an increase in training accuracy from 56% to 98% and a decrease in loss value from 1.9599 to 0.0730. Model evaluation using the Labeled Faces in the Wild (LFW) dataset achieved an accuracy of 96.13%, while validation accuracy reached 93.51%, indicating the model’s robustness under varying lighting conditions, facial expressions, and viewing angles. To enhance security, the system is equipped with an anti-spoofing module based on the Silent Face Anti-Spoofing library, capable of detecting fraudulent attempts such as the use of photos or videos. The user interface is implemented using Streamlit, enabling real-time attendance logging on devices equipped with a camera. The results show that the system is effective and secure for attendance recording and has the potential to be further developed into a scalable cloud-based service.
| Item Type: | Thesis (S1) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
|||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > T Technology (General) |
|||||||||
| Divisions: | 03-Fakultas Teknik 03-Fakultas Teknik > 55201-Jurusan Teknik Informatika |
|||||||||
| Depositing User: | Mr Fadel Najmi Adliansyah | |||||||||
| Date Deposited: | 04 Aug 2025 05:58 | |||||||||
| Last Modified: | 04 Aug 2025 05:58 | |||||||||
| URI: | http://eprints.untirta.ac.id/id/eprint/53460 |
Actions (login required)
 |
View Item |