Publication detail

Holistic Recognition of Low Quality License Plates by CNN using Track Annotated Data

ŠPAŇHEL, J. SOCHOR, J. JURÁNEK, R. HEROUT, A. MARŠÍK, L. ZEMČÍK, P.

Original Title

Holistic Recognition of Low Quality License Plates by CNN using Track Annotated Data

Type

conference paper

Language

English

Original Abstract

This work is focused on recognition of license plates in low resolution and low quality images. We present a methodology for collection of real world (non-synthetic) dataset of low quality license plate images with ground truth transcriptions. Our approach to the license plate recognition is based on a Convolutional Neural Network which holistically processes the whole image, avoiding segmentation of the license plate characters. Evaluation results on multiple datasets show that our method significantly outperforms other free and commercial solutions to license plate recognition on the low quality data. To enable further research of low quality license plate recognition, we make the datasets publicly available.

Keywords

holistic license plate recognition, convolutional neural network, low resolution, low quality

Authors

ŠPAŇHEL, J.; SOCHOR, J.; JURÁNEK, R.; HEROUT, A.; MARŠÍK, L.; ZEMČÍK, P.

Released

3. 8. 2017

Publisher

IEEE Computer Society

Location

Lecce

ISBN

978-1-5386-2939-0

Book

International Workshop on Traffic and Street Surveillance for Safety and Security (AVSS 2017)

Pages from

1

Pages to

6

Pages count

6

URL

http://ieeexplore.ieee.org/abstract/document/8078501/

BibTex

@inproceedings{BUT144463,
  author="Jakub {Špaňhel} and Jakub {Sochor} and Roman {Juránek} and Adam {Herout} and Lukáš {Maršík} and Pavel {Zemčík}",
  title="Holistic Recognition of Low Quality License Plates by CNN using Track Annotated Data",
  booktitle="International Workshop on Traffic and Street Surveillance for Safety and Security (AVSS 2017)",
  year="2017",
  pages="1--6",
  publisher="IEEE Computer Society",
  address="Lecce",
  doi="10.1109/AVSS.2017.8078501",
  isbn="978-1-5386-2939-0",
  url="http://ieeexplore.ieee.org/abstract/document/8078501/"
}