A Fast Method for Discovering Vehicle License Plate From Image with Complicated Background

Author

Guo, Fenglin ; Guan, Shu-an

Author_Institution

Dept. of Comput. & Inf. Eng., Wuhan Polytech. Univ., Wuhan, China

Volume

3

fYear

2009

fDate

4-5 July 2009

Firstpage

314

Lastpage

317

Abstract

The paper proposed a fast method on discovering vehicle license plate position in image with very complicated background and bad shooting condition. Firstly, we deduced a new RGB threshold formula by limiting the hue change range. Secondly, we provided a fast method to detect license plate position according to the characteristics of colors of license plate. Finally, we performed a detection experiment with 416 images containing license plate and the result showed that almost all license plates could be discovered quickly with our method from the images with even quit complicated background.

Keywords

image colour analysis; image recognition; vehicles; RGB threshold formula; color characteristics; complicated background; image detection; vehicle license plate recognition; Automotive engineering; Character recognition; Humans; Image color analysis; Image resolution; Image segmentation; Licenses; Pixel; Signal resolution; Vehicles; RGB thresholds; color sample set; license plate recognition; vehicle license plate;

fLanguage

English

Publisher

ieee

Conference_Titel

Environmental Science and Information Application Technology, 2009. ESIAT 2009. International Conference on

Conference_Location

Wuhan

Print_ISBN

978-0-7695-3682-8

Type

conf

DOI

10.1109/ESIAT.2009.443

Filename

5199698