International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 21 - Number 10 |
Year of Publication: 2011 |
Authors: Aicha Eutamene, Hacene Belhadef, M. K. Kholladi |
10.5120/2623-3648 |
Aicha Eutamene, Hacene Belhadef, M. K. Kholladi . Ontology based New Approach for Character Recognition. International Journal of Computer Applications. 21, 10 ( May 2011), 40-44. DOI=10.5120/2623-3648
The domain of pattern recognition and especially the character recognition is an area rich in term of scientific production, despite the abundance of work in this area but there are always anomalies. In this paper, we present a new approach for character recognition based on ontology, this last is carefully created with a domain expert, it contains a set of concepts and relationships, where each concept represents a grapheme, it is a feature extracted by an extraction module of a recognition system. Relationships between concepts are type spatial; describing the different possible relationships can be used between the graphemes, thus the characters in a document written in Latin alphabet. Our ontology is generic and can support other languages by enriching it by new specific spatial relationships.