Mean shift for accurate license plate localization

Wenjing Jia; Huaifeng Zhang; Xiangjian He; Massimo Piccardi

doi:10.1109/ITSC.2005.1520110

Mean shift for accurate license plate localization

Wenjing Jia, Huaifeng Zhang, Xiangjian He, Massimo Piccardi

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

70 Citations (Scopus)

Abstract

This paper presents a region-based algorithm for accurate license plate localization, where mean shift is utilized to filter and segment color vehicle images into candidate regions. Three features are extracted in order to decide whether a candidate region represents a real license plate, namely, rectangularity, aspect ratio, and edge density. Then, the Mahalanobis classifier is used with respect to above three features to classify license plate regions and non-license plate regions. Experimental results show that the proposed algorithm produces high robustness and accuracy.

Original language	English
Title of host publication	ITSC`05
Subtitle of host publication	2005 IEEE Intelligent Conference on Transportation Systems, Proceedings
Pages	566-571
Number of pages	6
DOIs	https://doi.org/10.1109/ITSC.2005.1520110
Publication status	Published - 2005
Externally published	Yes
Event	8th International IEEE Conference on Intelligent Transportation Systems - Vienna, Austria Duration: 13 Sept 2005 → 16 Sept 2005

Publication series

Name	IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC
Volume	2005

Conference

Conference	8th International IEEE Conference on Intelligent Transportation Systems
Country/Territory	Austria
City	Vienna
Period	13/09/05 → 16/09/05

ASJC Scopus subject areas

Automotive Engineering
Mechanical Engineering
Computer Science Applications

Access to Document

10.1109/ITSC.2005.1520110

Cite this

@inproceedings{6760959868aa4621bd3d52dc27f6b57d,

title = "Mean shift for accurate license plate localization",

abstract = "This paper presents a region-based algorithm for accurate license plate localization, where mean shift is utilized to filter and segment color vehicle images into candidate regions. Three features are extracted in order to decide whether a candidate region represents a real license plate, namely, rectangularity, aspect ratio, and edge density. Then, the Mahalanobis classifier is used with respect to above three features to classify license plate regions and non-license plate regions. Experimental results show that the proposed algorithm produces high robustness and accuracy.",

author = "Wenjing Jia and Huaifeng Zhang and Xiangjian He and Massimo Piccardi",

year = "2005",

doi = "10.1109/ITSC.2005.1520110",

language = "English",

isbn = "0780392159",

series = "IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC",

pages = "566--571",

booktitle = "ITSC`05",

note = "8th International IEEE Conference on Intelligent Transportation Systems ; Conference date: 13-09-2005 Through 16-09-2005",

}

Jia, W, Zhang, H, He, X & Piccardi, M 2005, Mean shift for accurate license plate localization. in ITSC`05: 2005 IEEE Intelligent Conference on Transportation Systems, Proceedings., 1520110, IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC, vol. 2005, pp. 566-571, 8th International IEEE Conference on Intelligent Transportation Systems, Vienna, Austria, 13/09/05. https://doi.org/10.1109/ITSC.2005.1520110

Mean shift for accurate license plate localization. / Jia, Wenjing; Zhang, Huaifeng; He, Xiangjian et al.
ITSC`05: 2005 IEEE Intelligent Conference on Transportation Systems, Proceedings. 2005. p. 566-571 1520110 (IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC; Vol. 2005).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Mean shift for accurate license plate localization

AU - Jia, Wenjing

AU - Zhang, Huaifeng

AU - He, Xiangjian

AU - Piccardi, Massimo

PY - 2005

Y1 - 2005

N2 - This paper presents a region-based algorithm for accurate license plate localization, where mean shift is utilized to filter and segment color vehicle images into candidate regions. Three features are extracted in order to decide whether a candidate region represents a real license plate, namely, rectangularity, aspect ratio, and edge density. Then, the Mahalanobis classifier is used with respect to above three features to classify license plate regions and non-license plate regions. Experimental results show that the proposed algorithm produces high robustness and accuracy.

AB - This paper presents a region-based algorithm for accurate license plate localization, where mean shift is utilized to filter and segment color vehicle images into candidate regions. Three features are extracted in order to decide whether a candidate region represents a real license plate, namely, rectangularity, aspect ratio, and edge density. Then, the Mahalanobis classifier is used with respect to above three features to classify license plate regions and non-license plate regions. Experimental results show that the proposed algorithm produces high robustness and accuracy.

UR - http://www.scopus.com/inward/record.url?scp=33747395185&partnerID=8YFLogxK

U2 - 10.1109/ITSC.2005.1520110

DO - 10.1109/ITSC.2005.1520110

M3 - Conference contribution

AN - SCOPUS:33747395185

SN - 0780392159

SN - 9780780392151

T3 - IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC

SP - 566

EP - 571

BT - ITSC`05

T2 - 8th International IEEE Conference on Intelligent Transportation Systems

Y2 - 13 September 2005 through 16 September 2005

ER -

Mean shift for accurate license plate localization

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this