Title :
Alphanumerical character recognition based on morphological analysis
Author :
Aljoša Dudarin;Zdenko Kovačić
Author_Institution :
Faculty of Electrical Engineering and Computing, University of Zagreb, Unska 3, 10000 Zagreb, Croatia
Abstract :
In this paper, we describe an alphabet character recognition method based on the morphological analysis of characters. The method can be also applied to numerals. Recognition is successful regardless of the character´s width, size and color. The constraint is that the method works only with standard upright fonts. The resulting character recognition algorithm is very simple and suitable for implementation on low-cost hardware. The algorithm has been successfully tested on the example of a car license plate recognition.
Keywords :
"Character recognition","Cavity resonators","Pixel","Licenses","Feature extraction","Shape","Decision trees"
Conference_Titel :
IECON 2010 - 36th Annual Conference on IEEE Industrial Electronics Society
Print_ISBN :
978-1-4244-5225-5
DOI :
10.1109/IECON.2010.5675515