Automatic Indonesian License Plate Recognition with YOLO as Object Detector

• Author(s): Sekar Larasati Muslimah ; Nevindra Ibnazhifi ; Kimberley Blessinda
• Paper ID: 1703068
• Page: 184-190
• Published Date: 28-12-2021
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 5 Issue 6 December-2021

Abstract

This paper created a license plate and vehicle number detection system to facilitate parking management and regularity. This system aims to change the traditional way of doing manual recording into computational automation of vehicle numbers. In this study, the Indonesian License Plate Recognition System consists of 2 processes, namely Object Plate Detection and Digit Number Recognition. The construction of the Object Plate Detector is carried out using the YOLOv4 transfer learning model. Meanwhile, Digit Number Recognition is built using the YOLOv5 transfer learning model. In this study, the YOLO model is used which is the current state-of-the-art object detector. With an accuracy of 89% for localization and detection of car plates and an accuracy of 87% for classifying characters on vehicle number plates.

Keywords

License Plate Recognition, Object Plate Detection, Digit Number Recognition, YOLOv4, YOLOv5

Citations

IRE Journals:
Sekar Larasati Muslimah , Nevindra Ibnazhifi , Kimberley Blessinda "Automatic Indonesian License Plate Recognition with YOLO as Object Detector" Iconic Research And Engineering Journals Volume 5 Issue 6 2021 Page 184-190

IEEE:
Sekar Larasati Muslimah , Nevindra Ibnazhifi , Kimberley Blessinda "Automatic Indonesian License Plate Recognition with YOLO as Object Detector" Iconic Research And Engineering Journals, 5(6)