Perancangan dan Analisis Optical Character Recognition (OCR) Untuk Mencocokan Pelat Kendaraan Pada Data Base

Alan Suryawinata, Dede Irawan Saputra

Abstract


This study aims to design a system that can detect license plates of motor vehicles that are located at the parking location of PT Tirta Gemah Ripah. The vehicle license plate detection system is designed using the OCR method which combines the haar cascade algorithm, Gaussian filter, and K-Nearest Neighbor (KNN) and equates it to the vehicle database. KNN will read the characters on the plate by using a training dataset of 500 characters. System accuracy testing uses several methods including differences in the number of training datasets, differences in the value of K, the elimination of the haar cascade algorithm, and the removal of the gaussian filter algorithm to find the best condition of the system when reading characters on the plates.
The haar cascade and Gaussian filter algorithm are algorithms that make the readings focused on the plate only and eliminate the object of interference in the image. Research data shows the amount of training data and K values for the KNN algorithm are very important with the results of the system data, the more training data, and the K values, the increase will also increase. The final results of the study using the value of K = 5 have 90.26% for the calculation of the accuracy of each character and 60% for the calculation of the accuracy of the entire plate.


Keywords


Plate recognition, Optical Character Recognition, KNN

References


D. Gunawan, W. Rohimah, and R. F. Rahmat, “Automatic Number Plate Recognition for Indonesian License Plate by Using K-Nearest Neighbor Algorithm,” in IOP Conference Series: Materials Science and Engineering, 2019, vol. 648, no. 1, doi: 10.1088/1757-899X/648/1/012011.

Y. Mansyur, “Optical Character Recognition Untuk Deteksi Pelat Mobil Dan Motor Kendaraan Pada Kampus Teknik Gowa,” Skripsi. Univ. Hasanuddin, 2018.

A. Budianto, T. B. Adji, and R. Hartanto, “Deteksi Nomor Kendaraan Dengan Metode Connected Component Dan Svm,” J. Teknol. Inf. Magister Darmajaya, vol. 1, no. 01, pp. 106–117, 2015.

P. P. Khare, R. Dudhe, A. Chungade, and A. Naykinde, “Advanced License Number Plate Recognition System,” Int. J. Eng. Res. Technol., vol. 3, no. 06, pp. 1–9, 2015, [Online]. Available: www.ijert.org.

Y. Puspito, F. X. A. Setyawan, and H. Fitriawan, “Deteksi Posisi Plat Nomor Kendaraan Menggunakan Metode Transformasi Hough Dan Hit Or Miss,” Electr. - J. Rekayasa dan Teknol. Elektro, vol. 12, no. 3, 2018.

A. P. Wahyu and Suhendri, “Peningkatan Sistem Keamanan Parkir dengan Teknologi Artificial Intelligence Imaging,” J. Inf. Technol., vol. 01, no. 02, pp. 39–42, 2019.

K. Kusumawati and D. W. Cahyadi, “Penerapan Teknologi Optical Character Recognition Untuk Mendeteksi Plat Nomor Kendaraan,” Pros. Semin. Nas. Inov. Teknol., pp. 12–20, 2017.

S. Wira, “Aplikasi Pendeteksi Plat Nomor Negara Indonesia Menggunakan Android Studio,” Nasional, no. December, p. 6, 2017, [Online]. Available: https://www.researchgate.net/publication/321797435_Aplikasi_Pendeteksi_Plat_Nomor_Negara_Indonesia_Menggunakan_Android_Studio.

N. A. Tauchid, “Analisis Performansi Metode KNN (K-Nearest Neighbor) Untuk Pengenalan Karakter Pada Plat Nomor Kendaraan di Raspberry Pi,” Telkom University, 2015.


Refbacks

  • There are currently no refbacks.


Copyright (c) 2021 CC BY SA

TELNECT https://ejournal.upi.edu/index.php/TELNECT/index di lisensikan di bawah Creative Commons Attribution-ShareAlike 4.0 International License.