Amilzan Fadlan Radiansyah, . (2026) SISTEM VERIFIKASI GANDA AKSES PARKIR BERBASIS DETEKSI PLAT NOMOR DAN PENGENALAN PENGEMUDI MENGGUNAKAN YOLO DAN OCR PADA RASPBERRY PI 4. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (803kB) |
|
Text
AWAL.pdf Download (1MB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (689kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (2MB) |
|
Text
BAB 5.pdf Download (758kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (607kB) |
|
Text
DAFTAR RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (197kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (3MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (24MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (1MB) |
Abstract
This study designs and implements a dual-verification parking access system based on computer vision technology to address the security vulnerabilities of conventional physical tickets. The proposed system integrates real-time license plate detection and driver face recognition by adopting a distributed Edge-to-Server architecture. A Raspberry Pi 4 Model B is configured as an on-site edge terminal for digital image acquisition and physical actuator control, while a laptop serves as the central server to handle the heavy computational load of deep learning models. License plate localization utilizes the YOLOv8s model trained on 1143 images, achieving a Precision of 98.80%, Recall of 95.38%, and mean Average Precision (mAP@0.5) of 97.20%. Alphanumeric characters are extracted using Optical Character Recognition (OCR), while driver face verification is performed via the FaceNet CNN architecture by calculating Euclidean distance (d ≤ 1.0) against a serialized database (.pkl). Full integration testing with 80 transaction samples demonstrates an Accuracy of 91.25%, Recall of 91.25%, F1-Score of 95.42%, and absolute security with a Precision of 100.00% (False Acceptance Rate = 0%). The system records an average end-to-end latency of under 2.35 seconds while maintaining the Raspberry Pi 4 CPU temperature within a stable range of 47.35°C to 53.79°C. The implementation of a trapezoidal acceleration profile on the Arduino Nano firmware successfully controls the NEMA 17 stepper motor to rotate the parking gate by 90° smoothly in 1.152 seconds without mechanical skip-steps. This distributed system is proven to be secure, responsive, and feasible for deployment in commercial smart parking infrastructures.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210314006] [Pembimbing 1: Ni Putu Devira Ayu Martini] [Pembimbing 2: Achmad Zuchriadi P] [Penguji 1: Didit Widiyanto] [Penguji 2: Andre Suwardana Adiwidya] |
| Uncontrolled Keywords: | Raspberry Pi 4, YOLOv8s, OCR, FaceNet, Edge-to-Server, Dual Verification. |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Teknik > Program Studi Teknik Elektro (S1) |
| Depositing User: | AMILZAN FADLAN RADIANSYAH |
| Date Deposited: | 28 Aug 2026 02:53 |
| Last Modified: | 28 Aug 2026 02:53 |
| URI: | http://repository.upnvj.ac.id/id/eprint/52722 |
Actions (login required)
 |
View Item |