Andika Febriyansyah, . (2026) OPTIMALISASI KINERJA SISTEM PENGENALAN SIDIK JARI PADA OPTICAL FINGERPRINT SENSOR MENGGUNAKAN SIAMESE NEURAL NETWORK (SNN). Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
This is the latest version of this item.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (52kB) |
|
Text
AWAL.pdf Download (997kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (94kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (685kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 5.pdf Download (82kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (159kB) |
|
Text
DAFTAR RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (16kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (18MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (547kB) |
Abstract
The AS608 optical fingerprint sensor exhibits limited performance under non-ideal finger surface conditions, such as dry, wet, and oily, because its internal matching algorithm relies heavily on the clarity of ridge patterns. This study optimizes the fingerprint recognition performance of the AS608 sensor using a Siamese Neural Network (SNN) with an EfficientNet-B0 architecture and BatchHard Triplet Loss, implemented on a website integrated with an ESP32-WROOM-32U microcontroller. The system was evaluated using an open-set identification scheme, comparing each sample against all database templates to find the candidate with the highest matching score, tested on 10 respondents under four finger surface conditions: normal, dry, wet, and oily. During validation, the SNN model achieved an Equal Error Rate (EER) of 5.2%, with Accuracy, Precision, Recall, and F1-Score each reaching 94.8%, and an AUC of 98.76%. During system testing, the SNN method achieved an average accuracy of 87.5%, a 32.1% improvement over the Internal Sensor method, reaching only 66.25%, with the average False Rejection Rate (FRR) decreasing from 67.5% to 25%, a 62.9% reduction, while False Acceptance Rate (FAR) remained stable at 0% for both methods. These results demonstrate that the SNN method improves fingerprint recognition performance, particularly under non-ideal finger surface conditions.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210314042] [Pembimbing 1: Andre Suwardana Adiwidya] [Pembimbing 2: Ni Putu Devira Ayu Martini] [Penguji 1: Achmad Zuchriadi P] [Penguji 2: Ayu Mika Sherila] |
| Uncontrolled Keywords: | EfficientNet-B0, Fingerprint, Identification, Optical Fingerprint Sensor, Siamese Neural Network |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software Q Science > QH Natural history > QH301 Biology T Technology > T Technology (General) |
| Divisions: | Fakultas Teknik > Program Studi Teknik Elektro (S1) |
| Depositing User: | ANDIKA FEBRIYANSYAH |
| Date Deposited: | 28 Aug 2026 03:11 |
| Last Modified: | 28 Aug 2026 03:11 |
| URI: | http://repository.upnvj.ac.id/id/eprint/52685 |
Available Versions of this Item
- OPTIMALISASI KINERJA SISTEM PENGENALAN SIDIK JARI PADA OPTICAL FINGERPRINT SENSOR MENGGUNAKAN SIAMESE NEURAL NETWORK (SNN). (deposited 28 Aug 2026 03:11) [Currently Displayed]
Actions (login required)
 |
View Item |