Parametric and structural adaptation of text labels recognition systems
E.M. Voskresensky, V.A.Tsarev

Institute of Management and Information Technologies (branch)
 of the St.-Petersburg State Polytechnical University

Full text of article: Russian language.

Abstract:
In article a new methods of multilevel text label recognition systems parametric and structural adaptation is offered. Such systems is described as hierarchical composition of algorithms. Methods allow adapting parameters of algorithms and parameters of decision-making subsystem of multilevel text label recognition systems. Method of multilevel text label recognition systems efficiency increase with their structural adaptation without necessity of algorithms modification is offered.

Key words:
recognition of symbols, recognition systems adaptation.

References:

  1. Voskresensky, E.M. Method of solution lists cut algorithms in text label recognition systems parametric optimization / E.M. Voskresensky, V.A. Tsarev // Computer Optics. – 2008. – V. 33, N 2. – P. 202-209. – (in Russian).
  2. Voskresensky, E.M. Method of estimation of text labels recognition systems efficiency parameters with use of likelihood characteristics tree / E.M. Voskresensky, V.A. Tsa­rev // Computer Optics. – 2008. – V. 32, N 3. – P. 283-290. – (in Russian).

© 2009, ИСОИ РАН
Россия, 443001, Самара, ул. Молодогвардейская, 151; электронная почта: ko@smr.ru ; тел: +7 (846 2) 332-56-22, факс: +7 (846 2) 332-56-20