Pengenalan Karakter Plat Nomor Kendaraan Menggunakan Metode Optical Character Recognition (OCR)

Bayu Gilang Sandjaya, Adlian Jefiza

Abstract

Abstrak : 

Artikel penelitian ini bertujuan untuk mengevaluasi akurasi sistem Optical Character Recognition (OCR) menggunakan metode YOLO V5 digunakan dalam pengenalan karakter plat nomor kendaraan dengan mempertimbangkan faktor-faktor seperti kondisi fisik plat, pencahayaan, dan jarak pengambilan gambar. Melalui analisis terhadap 14 data plat nomor, ditemukan bahwa plat dalam kondisi baik menunjukkan akurasi tinggi, mencapai 100% dalam banyak kasus, sedangkan plat yang cacat mengalami penurunan akurasi yang signifikan, dengan nilai terendah mencapai 0%. Penelitian menggunakan 5632 gambar plat motor dengan wana kuning, merah, putih, hitam dan hijau. Hasil pengujian juga mengindikasikan bahwa pencahayaan yang baik, terutama pada siang hari, sangat mendukung keberhasilan deteksi, sementara kondisi malam cenderung menurunkan akurasi, khususnya untuk plat yang cacat. Jarak pengambilan gambar juga berpengaruh, di mana jarak yang lebih jauh meningkatkan kemungkinan kesalahan dalam pembacaan karakter. Dari analisis ini, disarankan agar sistem OCR dapat dikembangkan lebih lanjut melalui peningkatan perangkat, penambahan fitur, dan pelatihan yang lebih mendalam dengan kumpulan data yang bervariasi.

================================================

Abstract : 

This research article aims to evaluate the accuracy of the Optical Character Recognition (OCR) system using the YOLO V5 method used in vehicle license plate character recognition by considering factors such as the physical condition of the plate, lighting, and shooting distance. Through an analysis of 14 license plate data, it was found that plates in good condition showed high accuracy, reaching 100% in most cases, while defective plates experienced a significant decrease in accuracy, with the lowest value reaching 0%. The study used 5632 images of motorcycle license plates with yellow, red, white, black and green colors. The test results also indicated that good lighting, especially during the day, greatly supports successful detection, while night conditions tend to decrease accuracy, especially for defective plates. The shooting distance also has an effect, where a longer distance increases the possibility of errors in character reading. From this analysis, it is suggested that the OCR system can be further developed through device improvements, additional features, and more in-depth training with varied data sets.

Keywords

Optical Character Recognition (OCR), Detection accuracy, Image processing, Vehicle license plate

Full Text:

PDF

References

[1] R. U. Amaliah and I. Refindo, “Faktor – Faktor Yang Berhubungan Dengan Kejadian Kecelakaan Pada Pengendara Ojek Online Batam,” J. Kesehat. Ibnu Sina J-KIS, vol. 1, no. 02, pp. 11–19, Aug. 2020, doi: 10.36352/j-kis.v1i02.107.

[2] D. E. Kurniawan and S. Fani, “Perancangan Sistem Kamera Pengawas Berbasis Perangkat Bergerak Menggunakan Raspberry PI,” J. Ilm. Teknol. Infomasi Terap., vol. 3, no. 2, Apr. 2017, doi: 10.33197/jitter.vol3.iss2.2017.130.

[3] I. Abdul Kadir, “Optimalisasi Penerapan Standar Pelayanan Minimal (SPM) SATPOL PP Maluku Utara: Menegakkan Ketentraman dan Ketertiban Umum Serta Melindungi Masyarakat,” Sci. J. Multi Discip. Sci., vol. 2, no. 2, pp. 72–83, Dec. 2023, doi: 10.62394/scientia.v2i2.68.

[4] Y. Darmi, M. Fajri Sepriansyah, Y. Darnita, and P. Pahrizal, “Penerapan Metode Optical Character Recognition (OCR) Untuk Mengidentifikasi Teks Pada Identitas Dokumen Surat Izin Mengemudi (SIM),” JATI J. Mhs. Tek. Inform., vol. 9, no. 4, pp. 5992–5998, May 2025, doi: 10.36040/jati.v9i4.13987.

[5] Wanda Hamidah, T. S. B. Irawan, N. A. P. Hasbullah, and A. B. Kaswar, “Deteksi Nominal Uang Kertas Menggunakan OCR (Optical Character Recognition),” Techno Xplore J. Ilmu Komput. Dan Teknol. Inf., vol. 7, no. 2, pp. 72–76, Oct. 2022, doi: 10.36805/technoxplore.v7i2.2123.

[6] J. Valentino and Y. A. Susetyo, “Analisis Perbandingan Optical Character Recognition Google Vision dengan Microsoft Computer Vision pada Pembacaan KTP-el,” J. JTIK J. Teknol. Inf. Dan Komun., vol. 7, no. 4, pp. 552–561, Oct. 2023, doi: 10.35870/jtik.v7i4.1046.

[7] F. Kindarya et al., “Penerapan Aplikasi Klasifikasi Hukum Tajwid Menggunakan Image Processing,” El-Mujtama J. Pengabdi. Masy., vol. 4, no. 2, Feb. 2024, doi: 10.47467/elmujtama.v4i2.1930.

[8] I. Saitov and A. Filchenkov, “CIS Multilingual License Plate Detection and Recognition Based on Convolutional and Transformer Neural Networks,” Procedia Comput. Sci., vol. 229, pp. 149–157, 2023, doi: 10.1016/j.procs.2023.12.016.

[9] M. Hanum, “Implementasi Teknik Embossing pada Pengenalan Plat Kendaraan untuk Identifikasi Otomatis Berbasis OpenCV,” JoMMiT J. Multi Media Dan IT, vol. 8, no. 1, pp. 062–068, Jul. 2024, doi: 10.46961/jommit.v8i1.1361.

[10] R. Rismanto, A. Prasetyo, and D. A. Irawati, “Optimalisasi Image Thresholding pada Optical Character Recognition Pada Sistem Digitalisasi dan Pencarian Dokumen,” PETIR, vol. 13, no. 1, pp. 1–11, Mar. 2020, doi: 10.33322/petir.v13i1.659.

[11] I. Setiawan, W. Dewanta, H. A. Nugroho, and H. Supriyono, “Pengolah Citra Dengan Metode Thresholding Dengan Matlab R2014A,” J. MEDIA INFOTAMA, vol. 15, no. 2, Oct. 2019, doi: 10.37676/jmi.v15i2.868.

[12] H. Xu, Q. Li, and J. Chen, “Highlight Removal from A Single Grayscale Image Using Attentive GAN,” Appl. Artif. Intell., vol. 36, no. 1, p. 1988441, Dec. 2022, doi: 10.1080/08839514.2021.1988441.

[13] G. De Carvalho Oliveira, C. C. S. Machado, D. K. Inácio, J. F. D. Silveira Petruci, and S. G. Silva, “RGB color sensor for colorimetric determinations: Evaluation and quantitative analysis of colored liquid samples,” Talanta, vol. 241, p. 123244, May 2022, doi: 10.1016/j.talanta.2022.123244.

[14] Z. Luo, D. Shi, X. Shen, J. Ji, and W.-S. Gan, “GFANC-Kalman: Generative Fixed-Filter Active Noise Control With CNN-Kalman Filtering,” IEEE Signal Process. Lett., vol. 31, pp. 276–280, 2024, doi: 10.1109/LSP.2023.3334695.

[15] M. M. Hamid, F. Fathi Hammad, and N. Hmad, “Removing the Impulse Noise from Grayscaled and Colored Digital Images Using Fuzzy Image Filtering,” in 2021 IEEE 1st International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering MI-STA, Tripoli, Libya: IEEE, May 2021, pp. 711–716. doi: 10.1109/MI-STA52233.2021.9464371.

DOI: https://doi.org/10.20961/ijai.v10i1.108640

Refbacks

  • There are currently no refbacks.