Abstract—Today the necessity of using smart systems which are able to diagnose the cars plague correctly from pictures is felt more than ever by expanding the number of vehicles. In License Plate Recognition systems, detecting plaque location is considered the most important phase so that the correct rate of it, will affect all subsequent phases. There are different images with different resolutions and sever changes in the size of images which taken through being far or near to the camera, it makes many existing methods face problems or longer processing. In this paper a new method is suggested based on simulation concepts which used in spectra analysis for accurate and fast position detection by using edge concepts and morphological operators in image processing. The suggested method is a compound method and fast enough and it extracts the plague location from vehicle image correctly. The algorithm is implemented in MATLAB and the results obtained agreed with theoretical predictions. The proposed method has 93.3 % of correct detection rate and 150 ms of processing time.
Index Terms—LPR, Edge detection, Morphological operators, Spectral analysis.
All Authors are with Department of Computer Engineering and Information Technology, Shiraz University of Technology, Shiraz, Iran (e-mail: mehdirahmanian @ ymail.com; reza.javidan@gmail.com; keshtgari @sutech.ac.ir).
[PDF]
Cite: Mehdi Rahmanian, Reza Javidan, Member, IACSIT, and Manijeh Keshtgary, "A New Method for Positioning Persian License Plates," International Journal of Engineering and Technology vol. 3, no. 6, pp. 669-674, 2011.
