Title: Automatic thresholding of license plate

Authors: Khalid Aboura

Addresses: Centre for Built Infrastructure Research, Faculty of Engineering and Information Technology, University of Technology, Sydney

Abstract: Licence plate recognition (LPR) is a well studied problem first attempted in the 1970s. The image analysis part in LPR renders the problem hard to solve perfectly, leading to continued research. A defining step is the thresholding of the image to separate an object in the image from its background. While considerable effort has been devoted to LPR, thresholding is hardly ever mentioned, always assuming an effective one. A large number of thresholding methods exist. These methods perform differently in various image analysis problems but rather poorly if applied in an automatic fashion to the LPR problem. We introduce a new thresholding method that performs well in automatic LPR. The method uses colour-based histogram classification information. The approach leads to a 93.4% success rate and can be applied to the automatic recognition of symbols and characters.

Keywords: image thresholding; colour spectrum histogram; statistical methodology; linear regression; automation; licence plates; licence plate recognition; LPR; symbol recognition; character recognition; number plates.

DOI: 10.1504/IJAAC.2008.022178

International Journal of Automation and Control, 2008 Vol.2 No.2/3, pp.213 - 231

Published online: 22 Dec 2008 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article