Number of the records: 1
Multivariate structural Bernoulli mixtures for recognition of handwritten numerals
- 1.0410442 - UTIA-B 20000158 RIV US eng C - Conference Paper (international conference)
Grim, Jiří - Pudil, Pavel - Somol, Petr
Multivariate structural Bernoulli mixtures for recognition of handwritten numerals.
Los Alamitos: IEEE Computer Society, 2000. ISBN 0-7695-0750-6. In: Proceedings of the 15th International Conference on Pattern Recognition. - (Sanfeliu, A.; Villanueva, J.; Vanrell, M.), s. 585-589
[International Conference on Pattern Recognition /15./. Barcelona (ES), 03.09.2000-07.09.2000]
Grant - others:GA AV(CZ) IAA2075703; MŠMT(CZ) VS96063; MŠMT(CZ) ME 187
Program: IA
Institutional research plan: AV0Z1075907
Subject RIV: BB - Applied Statistics, Operational Research
As shown recently, the structural optimization of probabilistic neural networks can be included into EM algorithm by introducing a special type of mixtures. The method has been applied to recognize unconstrained handwritten numerals from the database of Concordia University in Montreal. In the present paper we discuss the possibility of a proper initialization of EM algorithm for estimating the class-conditional multivariate Bernoulli mixtures.
Permanent Link: http://hdl.handle.net/11104/0130531
Number of the records: 1