Bayes classification of online Arabic characters by Gibbs modelling of class conditional densities [r-libre/424]

Mezghani, Neila; Mitiche, Amar, & Cheriet, Mohamed (2008). Bayes classification of online Arabic characters by Gibbs modelling of class conditional densities. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30 (7), 1121-1131. https://doi.org/10.1109/TPAMI.2007.70753

File(s) available for this item:

PDF - Arabic-character.pdf
Content : Published Version
Restricted access

Request a copy

Item Type:	Journal Articles
Refereed:	Yes
Status:	Published
Abstract:	This study investigates Bayes classification of online Arabic characters using histograms of tangent differences and Gibbs modeling of the class-conditional probability density functions. The parameters of these Gibbs density functions are estimated following the Zhu et al. constrained maximum entropy formalism,originally introducedfor image and shape synthesis. We investigate two partition function estimation methods: one uses the training sample, and the other draws from a reference distribution. The efficiency of the corresponding Bayes decision methods, and of a combination of these, is shown in experiments using a database of 9,504 freely written samples by 22 writers. Comparisons to the nearest neighbor rule method and a Kohonen neural network method are provided.
Depositor:	Mezghani, Neila
Owner / Manager:	Neila Mezghani
Deposited:	24 Oct 2014 16:48
Last Modified:	30 Oct 2019 15:33

Actions (login required)

RÉVISER