ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 7. Vol. 24. 2018

DOI: 10.17587/it.24.481-486

N. O. Efimov, Graduate Student, e-mail: n.o.efimov@gmail.com, E. G. Zhiljakov, D. Sc., Professor, department of Information and telecommunication systems and technologies, Belgorod National Research University, Belgorod, 308000, Russian Federation

Handwritten Text Fragment Recognition. Typical Subband Intervals

The research of existing methods of identification scanning text images allow to draw a conclusion, what in general case problem solve only for printed text images. In case of handwritten text each solution is individual, and frequently need a long time education or operator inspection. In the article introduce new decision procedure of precedential identical fragment detection, which react on normalized Fourier transform differences in defined frequency intervals. Possibility of detection typical subband intervals for specified symbol is analyzed. Using typical subband intervals can appreciably decrease number of features, necessary for fragment description.
Keywords: image processing, image recognition, HWR, OCR, substrip method, subband method, typical subband intervals, decision procedure, measure of proximity, Fourier transform

P. 481–486

To the contents