Skip Ribbon Commands Skip to main content
Help (new window)
University of Balamand > Academics > Research > Seminars > Ramy El-Hajj

Cursive Handwriting Recognition: Combination of Multi-HMMs Classifiers Based on Oriented Fragmentation

SUBMITTED BY: Ramy El-Hajj
ENST Paris

ABSTRACT: A new system for off-line recognition of handwritten texts is presented. The proposed system is based on an analytical approach and is composed by a two-stages structure. In the first stage, a new strategy for features’ extraction, based on oriented fragmentation is implemented. The post-treatment stage consists of a combination of Multi-HMMs Classifiers. The results of performance’s evaluation of a reference system are presented. And the enhancement of the recognition performance brought by the application of the new approach is shown.
facebook    LinkedIn    youTube    Social Media
University of Balamand,
Balamand Al Kurah,
Lebanon

Tel:  +961-6-930250
Fax: +961-6-930278