LICENSE-PLATE LOCALIZATION BY USING VECTOR
QUANTIZATION
Stefano Rovetta, Rodolfo Zunino (DIBE - Genoa University) 

In: PROOCEEDINGS of the
1999 IEEE International Conference On
ACOUSTICS, SPEECH, AND SIGNAL PROCESSING

March 15-19, 1999
Civic Plaza, Hyatt Regency - Phoenix, Arizona 

vol. 2, pp. 1113-1116 



A B S T R A C T 
---------------
The paper describes a novel approach using Vector Quantization (VQ) to process vehicle
images for automated identification. The VQ-based method yields superior quality in picture
compression for archival purposes, and, at the same time, supports the localization of text
regions in the image effectively. As opposed to standard approaches, VQ encoding gives some
hint about the contents of image regions; such information is exploited to boost localization
performance. The VQ system may be trained empirically from examples; this provides
adaptiveness and on-field tuning facility. The approach has been tested in a real application and
included satisfactorily into a complete system for vehicle identification.