A Novel Accurate Localization Method of Vehicle-logo

DOI

10.2991/icmmita-15.2015.295How to use a DOI?

Keywords

Vehicle-logo; Accurate localization; Gradient angle histogram analysis; Local self-adaptive threshold.

Abstract

Aiming at the problem lied in existed methods which cannot suppress grille background completely, a novel method of vehicle-logo accurate localization was proposed. Firstly color invariants and prior information are used for fast localizing license plate in RGB color space directly; and then process fine localization of the vehicle-logo based on coarsely localized results. Background noises are secondarily suppressed based on gradient angle histogram analysis and local self-adaptive threshold treatment. Numerous experiments show that this method possesses not only high accuracy but also high robustness in localization, as compared to classic template matching. Besides, the proposed method satisfies the requirement of real-time performance.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jin Liu
AU  - De Cai
AU  - Zhonghan Shi
AU  - Jiahao Deng
PY  - 2015/11
DA  - 2015/11
TI  - A Novel Accurate Localization Method of Vehicle-logo
BT  - Proceedings of the 2015 3rd International Conference on Machinery, Materials and Information Technology Applications
PB  - Atlantis Press
SP  - 1593
EP  - 1598
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmita-15.2015.295
DO  - 10.2991/icmmita-15.2015.295
ID  - Liu2015/11
ER  -