№3 (46), 2018. INFORMATION TECHNOLOGY, COMPUTER SCIENCE AND CONTROL

Zelenskiy A.A., Pismenskova M.M.

The human action recognition method on complex-structured images and stochastic texture background

The paper a new approach to recognizing human actions using global descriptors as part of an intelligent video surveillance system is present. The method proposed in this paper is based on the mathematical apparatus of stochastic geometry. It allows to generate a large number of new, constructive features and to describe similar images, to increase the effectiveness of recognizing human actions on complex-structured images and a background in the form of stochastic textures.

Keywords: action recognition, video sequence, complex-structured images, stochastic textures.

[ Back to contents ]
[ Get the article ]