Deteksi dan Pengenalan Dua Variasi Plat Nomor Kendaraan Bermotor di Indonesia Dengan Variasi Waktu dan Pencahayaan Memanfaatkan YOLO V8 dan CNN
Abstract
ABSTRACT
Vehicle license plates are one of the vehicle that one of the identity in Indonesia, and these identity are crucial for distinguishing one vehicle from another. License plates detection technology has been advancing, but it is still challenging to read license plates under poor weather conditions and lighting. This research aims to improve the accuracy and speed of license plate detection using YOLO version 8 for license plate detection and CNN for image-to-text extractions. The dataset used consists of 3000 images captured under various conditions (different angles, distances, weather conditions, and lighting). There are two models resulting from the training data : one model with images resized to 640x640 pixels and another model using original-sized images. The accuracy obtained from the dataset is 96.7% with an F1-Score of 0.99 at 0.590, demonstrating that YOLO version 8 method is superior to previous versions.
Keywords: Vehicle Number Plate, YOLOV8, CNN.
ABSTRAK
Plat nomor kendaraan merupakan salah satu identitas kendaraan di Indonesia, identitas tersebut sangat penting untuk mengidentifikasi kendaraan satu dengan yang lainnya. Teknologi deteksi plat nomor sudah mulai berkembang, namun sangat sulit untuk membaca plat nomor dengan kondisi cuaca dan pencahayaan yang kurang. Penelitian ini bertujuan untuk meningkatkan akurasi dan kecepatan pada metode deteksi plat nomor, dengan menggunakan YOLO versi 8 sebagai deteksi plat nomor dan CNN untuk ekstrasi gambar menjadi teks. Dataset yang digunakan berjumlah 3000 gambar dengan berbagai kondisi pengambilan gambar (sudut kemiringan, jarak pengambilan gambar, kondisi cuaca, dan cahaya). Terdapat 2 model dari hasil data training, yaitu 1 model dengan gambar yang dikecilkan menjadi ukuran piksel 640x640 dan 1 model lainnya menggunakan gambar asli. Akurasi yang didapat dari dataset tersebut adalah 96.7% dengan F1-Score 0.99 at 0.590, hal ini membuktikan bahwa metode YOLO versi 8 lebih baik daripada versi sebelumnya.
Kata Kunci: Plat nomor kendaraan, YOLOv8, CNN.
Downloads
References
S. Sadya, “Kendaraan Di Indonesia Pada 2022,” Data Indonesia, 2023. Https://Dataindonesia.Id/Sektor-Riil/Detail/Polri-Catat-15251-Juta-Kendaraan-Di-Indonesia-Pada-2022 (Accessed Jul. 31, 2023).
F. Indaryanto, A. Nugroho, And A. F. Suni, “Aplikasi Penghitung Jarak Dan Jumlah Orang Berbasis Yolo Sebagai Protokol Kesehatan Covid-19,” Edu Komputika Journal, Vol. 8, No. 1, Pp. 31–38, Jun. 2021, Doi: 10.15294/Edukomputika.V8i1.47837.
Ihsanudin And Ahadi Ayu, "Penerapan Framework Flask Pada Machine Learning Dalam Memprediksi Umur Transformer" Jurnal Konvergensi, Vol 19 No 2, 2023.
N. E. Budiyanta, M. Mulyadi, And H. Tanudjaja, “Sistem Deteksi Kemurnian Beras Berbasis Computer Vision Dengan Pendekatan Algoritma Yolo,”Jurnal Informatika: Jurnal Pengembangan It, Vol. 6, No. 1, Pp. 51–55, Jan. 2021, Doi: 10.30591/Jpit.V6i1.2309.
A.E. Wijaya, W. Swastika, And O.H.Kelana, “Implementasi Transfer Learning Pada Convolutional Neural Network Untuk Diagnosis Covid-19 Dan Pneumonia Pada Citra X-Ray,” Sainsbertek Jurnal Ilmiah Sains & Teknologi, Vol. 2, No. 1, Pp. 10–15, Sep. 2021, Doi: 10.33479/Sb.V2i1.125.
K. Khairunnas, E. M. Yuniarno, And A. Zaini, “Pembuatan Modul Deteksi Objek Manusia Menggunakan Metode Yolo Untuk Mobile Robot,” Jurnal Teknik Its, 2021, [Online]. Available: Http://Ejurnal.Its.Ac.Id/Index.Php/Teknik/Article/View/61622
A. Verma, C. Kapoor And A. Sharma, "Web Application Implementation With Machine Learning," California, 2021.
I. Reezky And Hustinawati, “Yolo V5 Untuk Deteksi Nomor Kendaraan Di Dki Jakarta Yolo V5 For Vehicle Plate Detection In Dki Jakarta,” 2023. [Online]. Available: Www.Kaggle.Com
K. Khotimah, M. T. Alawiy, And B. M. Basuki, “Pendeteksi Plat Nomor Kendaraan Bermotor Berbasis Algoritma Yolo ( You Only Look Once ) Menggunakan Kamera Cctv,” Vol. 1, No. 1, Pp. 1–8, 2023.
B. Setiyono, D. A. Amini, And D. R. Sulistyaningrum, “Number Plate Recognition On Vehicle Using Yolo - Darknet,” J. Phys. Conf. Ser., Vol. 1821, No. 1, 2021, Doi: 10.1088/1742-6596/1821/1/012049.
M. Henrique Fonseca Afonso Et Al., “Vehicle And Plate Detection For Intelligent Transport Systems: Performance Evaluation Of Models Yolov5 And Yolov8,” No. August, 2023, Doi: 10.13140/Rg.2.2.11022.95042.
H. Shi And D. Zhao, “License Plate Recognition System Based On Improved Yolov5 And Gru,” Ieee Access, Vol. 11, No. January, Pp. 10429–10439, 2023, Doi: 10.1109/Access.2023.3240439.
Yanto, Y., Aziz, F., & Irmawati, I. (2023). Yolo-V8 Peningkatan Algoritma Untuk Deteksi Pemakaian Masker Wajah. Jati (Jurnal Mahasiswa Teknik Informatika),7(3), 1437–1444.
G. Zeng, “On The Confusion Matrix In Credit Scoring And Its Analytical Properties,” Commun Stat Theory Methods, Vol. 49, No. 9, Pp. 2080–2093, May 2020, Doi: 10.1080/03610926.2019.1568485.
M. Grandini, E. Bagli, And G. Visani, “Metrics For Multi-Class Classification: An Overview,” Aug. 2020.
Authors whose manuscript is published will approve the following provisions:
- The right to publication of all journal material published on the Konvergensi Teknologi Informasi & Komunikasi website is held by the editorial board with the author's knowledge (moral rights remain the property of the author).
- The formal legal provisions for access to digital articles of this electronic journal are subject to the terms of the Creative Commons Attribution-ShareAlike (CC BY-SA) license, which means Konvergensi Teknologi Informasi & Komunikasi reserves the right to store, modify the format, administer in database, maintain and publish articles without requesting permission from the Author as long as it keeps the Author's name as the owner of Copyright.
- Printed and electronic published manuscripts are open access for educational, research and library purposes. In addition to these objectives, the editorial board shall not be liable for violations of copyright law.
