I. A. Bagrova, A. A. Gritsaj, S. A. Ponovarev, S. V. Sorokin, D. A. Sytnik, “Selecting features for recognizing printed Cyrillic characters”, Vestnik TVGU. Ser. Prikl. Matem. [Herald of Tver State University. Ser. Appl. Math.], 2010, Issue 18,Pages <nobr>59

Signal and Image Processing

Selecting features for recognizing printed Cyrillic characters

I. A. Bagrova, A. A. Gritsaj, S. A. Ponovarev, S. V. Sorokin, D. A. Sytnik

Complex systems Ltd, Tver

Abstract: Four types of features for recognition of Cyrillic letters are analyzed. Algorithm for feature selection based on ideas of genetic optimization and clustering is proposed.

Keywords: optical character recognition, clustering, genetic optimization, feature selection.

UDC: 004.932.75’1

Received: 02.06.2010
Revised: 04.06.2010