21 BASED FEATURE EXTRACTION OF A NUMBER PLATE IMAGE
Sushmita Kumari, Sudha. S, K. Brindha
ABSTRACT
In this proposed paper a comprehensive survey on existing ANPR techniques by categorizing them according to the features used in each stage. Automatic car number plate recognition is a technology that enables a computer system to read automatically the number plate of the car. This proposed system is basically for Indian car number plates that consist of English characters and English digits. The ANPR system consists of four basic steps Plate Localization, Pre-processing, Classification and Feature Extraction. But in our System, we are taking noise-free number plates with font Arial, Times New Roman, and Verdana. We have assumed that we are getting segmented data for training of characters and digits. The 21 based feature extraction method (zone based) and minimum distance classifier algorithms are used for this Automatic number plate recognition System. The characters and digits are recognized by using its ASCII values.
[Full Text Article]