Musyafira, Nuzha (2020) Deteksi Posisi dan Pengenalan Plat Nomor Kendaraan Menggunakan Single Shot Detector dan Recurrent Neural Network pada Data Video. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
05111640000014-Undergraduate_Thesis.pdf - Accepted Version Download (5MB) | Preview |
Abstract
Setiap kendaraan mempunyai identitasnya masing-masing, dengan kata lain, plat nomor kendaraan. Identitas ini sering kali dipergunakan dalam sistem pengolahan parkir, pengamanan, dan sebagainya. Untuk membuat sistem ini andal, diperlukan adanya pengembangan sistem otomatis yang dapat mendeteksi dan mengidentifikasi plat nomor kendaraan. Sistem ini telah banyak dikenal sebagai License Plate Recognition (LPR). LPR menggunakan konsep deteksi posisi (lokalisasi) dan segmentasi untuk identifikasi plat kendaraan pada suatu citra. Hasil plat yang telah terdeteksi nantinya akan dikenali sebagai karakter-karakter yang merepresentasikan identitas kendaraan.
Dalam tugas akhir ini, akan dilakukan pengembangan LPR yang memungkinkan membaca masukan dari data video yang mengacu pada real-time processing. Dengan menggunakan metode lokalisasi Single Shot Detector, segmentasi Binary Image diikuti dengan Connected Component Labelling, dan pengenalan karakter Recurrent Neural Network, hasil dari penelitian ini menunjukkan akurasi sebesar 91.48% untuk lokalisasi plat, 82.69% untuk segmentasi, dan 94.94% untuk pengenalan karakter.
==================================================================================================================================
Each vehicle has its own identity, in other words, the vehicle number plate. This identity is often used in parking processing, security, and so on. To make this system, it is necessary to develop an automated system that can be used and supported by vehicle number plates. This system is known as License Plate Recognition (LPR). LPR uses the concept of position detection (localization) and segmentation to determine vehicle plates in images. The results of the verified plate will be recognized as characters that represent the vehicle's identity.
In this undergraduate thesis, the development of LPR will be made which allows reading input in the form of video data that refers to real-time processing. By using the Single Shot Detector localization method, Binary Image segmentation followed by Connected Component Labeling, and Recurrent Neural Network character recognition, the results of this study show an accuracy of 91.48% for plate localization, 82.69% for segmentation, and 94.94% for character recognition.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSIf 006.32 Mus d-1 |
| Uncontrolled Keywords: | License Plate Recognition |
| Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Nuzha Musyafira |
| Date Deposited: | 26 Apr 2023 08:51 |
| Last Modified: | 26 Apr 2023 08:51 |
| URI: | http://repository.its.ac.id/id/eprint/73473 |
Actions (login required)
 |
View Item |