A novel 2-row Indian LP character segmentation algorithm based on a Hybrid approach

Author

Vishwanath, N. ; Somasundaram, Sivanand ; Begum, A.R.J. ; Nallaperumal, N.K.

Author_Institution

Cape Inst. of Technol., Kanniyakumari, Levengipuram, India

fYear

2012

fDate

18-20 Dec. 2012

Firstpage

1

Lastpage

4

Abstract

A Hybrid approach involving new thresholding method for binarizing the LP plate area with the background combined with a Centre Weighted Median Filter-CWMF noise pre-processing algorithm to get a clear Indian LP area to be passed on to the new zone character segmentation algorithm to get a near perfect segmented LP characters output. The thresholding algorithm initially uses a minimum threshold to locate the plate area with the background and calculates an optimum threshold value by minimizing the variance within the LP plate area while maximizing the variance between the plate area and its background. The new zone segmentation algorithm enables multiple rows Indian License Plate segmentation.

Keywords

character recognition; computer vision; image segmentation; median filters; traffic engineering computing; 2-row Indian LP character segmentation algorithm; CWMF noise preprocessing algorithm; Indian license plate segmentation; LP plate area binarization; centre weighted median filter; computer vision; hybrid approach; near perfect segmented LP character; new zone character segmentation algorithm; optimum threshold value; plate area location; thresholding algorithm; thresholding method; License plate; segmentation; two rows;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence & Computing Research (ICCIC), 2012 IEEE International Conference on

Conference_Location

Coimbatore

Print_ISBN

978-1-4673-1342-1

Type

conf

DOI

10.1109/ICCIC.2012.6510319

Filename

6510319