Information technologies and computer sciences

DETERMINATION OF EFFICIENCY OF COLOR CHARACTERISTICS IN FRAME OF VIOLA–JONES ALGORITHM BASED ON PERSONS LOCATION

Authors

I. O. Marchenko
Sumy State University ROR
S. O. Petrov
Sumy State University ROR
N. V. Lysak
Vinnytsia National Technical University ROR

Keywords

Viola–Jones algorithm localization of image skin density ratio optimization of the geometric parameter cognition window

Abstract

The paper showed color characteristics of human skin effecting localization problems solved by Viola-Jones algorithm in scope of face image detecting.

There has been offered the approach to optimizing the geometrical parameters of the results of localization using an additional parameter – the skin density coefficient. Based on the training set (13225 input images) using iterative procedure there has been determined the optimum value of this coefficient and considered the average effectiveness of the process to reduce geometric parameters for a localized image. Which is significantly affects the size of the stored results and the results of further recognition.

360 162

How to Cite

[1]
“DETERMINATION OF EFFICIENCY OF COLOR CHARACTERISTICS IN FRAME OF VIOLA–JONES ALGORITHM BASED ON PERSONS LOCATION”, Вісник ВПІ, no. 1, pp. 108–114, Mar. 2016, Accessed: Oct. 11, 2026. Available: https://visnyk.vntu.edu.ua/index.php/visnyk/article/view/1885

Author Biographies

I. O. Marchenko, Sumy State University
Post-Graduate Student of the Chair of Computer Sciences
S. O. Petrov, Sumy State University
Cand. Sc. (Eng.), Assistant Professor of the Chair of Computer Sciences
N. V. Lysak, Vinnytsia National Technical University
Cand. Sc. (Eng.), Assistant Professor of the Chair of Management and Security of Information Systems

Most read articles by the same author(s)