1 April 1993 Morphological hand-printed character recognition by a skeleton-matching algorithm
Author Affiliations +
J. of Electronic Imaging, 2(2), (1993). doi:10.1117/12.143731
Abstract
We study the use of mathematical morphology for handprinted character recognition. Our approach uses the morphological skeleton transform as the shape descriptor. An efficient skeleton-matching algorithm, which renders the similarity between two skeletons as a distance measure, is employed. Based on this distance measure, a character is classified by a minimum distance classifier. The morphological skeleton transform contains complete shape information and is shown as a powerful descriptor for this class of shapes. We also study the pattern spectrum as a shape descriptor for handprinted characters. However, the pattern spectrum conveys only information about the shape/size distribution of a given object, which turns out to be not very efficient for hand-printed characters. Experimental results demonstrate the efficiency of the skeleton-based approach and the inadequacy of the pattern-spectrum-based approach.
Panagiotis E. Trahanias, Konstantinos Stathatos, Fotios Stamatelopoulos, Emmanuel Skordalakis, "Morphological hand-printed character recognition by a skeleton-matching algorithm," Journal of Electronic Imaging 2(2), (1 April 1993). http://dx.doi.org/10.1117/12.143731
JOURNAL ARTICLE
12 PAGES


SHARE
KEYWORDS
Distance measurement

Optical character recognition

Picosecond phenomena

Shape analysis

Shape memory alloys

Detection and tracking algorithms

Binary data

RELATED CONTENT


Back to Top