A License Plate Recognition System for Severe Tilt Angles Using Mask R-CNN

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the past few years, license plate recognition systems have been widely used in parking lots. In order to identify license plates easily, traditional license plate recognition systems used in the parking lot have a fixed light source and a shooting angle. For particularly tilting angles, such as license plate images taken with super wide-Angle lenses or fisheye lenses, the deformation of the license plate can be particularly severe, resulting in poor recognition of traditional license plate recognition systems. In this paper, we propose a three-stage license plate recognition system based on Mask R-CNN that can be used for various shooting angles and more oblique images. Experimental results show that the proposed architecture can identify license plates with bevel angles over 0∼60 degrees and achieve mAP rates of up to 91%. Compared with the approach using YOLOv2 model, the proposed method with Mask R-CNN has made significant progress in identifying characters that are inclined above 45 degrees. The experimental results also show that the proposed method is superior to other methods in the open Taiwan license plate data set (called AOLP data set).

Original languageEnglish
Title of host publicationProceedings - 2019 International Conference on Advanced Mechatronic Systems, ICAMechS 2019
PublisherIEEE Computer Society
Pages229-234
Number of pages6
ISBN (Electronic)9781728134802
DOIs
Publication statusPublished - 2019 Aug
Event2019 International Conference on Advanced Mechatronic Systems, ICAMechS 2019 - Kusatsu, Shiga, Japan
Duration: 2019 Aug 262019 Aug 28

Publication series

NameInternational Conference on Advanced Mechatronic Systems, ICAMechS
Volume2019-August
ISSN (Print)2325-0682
ISSN (Electronic)2325-0690

Conference

Conference2019 International Conference on Advanced Mechatronic Systems, ICAMechS 2019
CountryJapan
CityKusatsu, Shiga
Period19/8/2619/8/28

    Fingerprint

Keywords

  • Mask R-CNN
  • deep learning
  • license plate recognition systems

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Mechanical Engineering

Cite this

Lin, C. H., & Li, Y. (2019). A License Plate Recognition System for Severe Tilt Angles Using Mask R-CNN. In Proceedings - 2019 International Conference on Advanced Mechatronic Systems, ICAMechS 2019 (pp. 229-234). [8861691] (International Conference on Advanced Mechatronic Systems, ICAMechS; Vol. 2019-August). IEEE Computer Society. https://doi.org/10.1109/ICAMechS.2019.8861691