CLASSIFICATION OF IMAGES ON THE BASIS OF THE PROPERTIES OF INFORMATIVE REGIONS

Shaposhnikov D.,1 Podladchikova L.1 and Gao X.2

1A.B. Kogan Research Institute for Neurocybernetics,
Rostov State University, Rostov-on-Don, Russia
2School of Computing Science, Middlesex University
London, United Kingdom


Abstract

Algorithms for classification of the images of faces and road signs are presented. Estimation of facial aspects is based on the determination of mutual geometric relations between the most informative regions (MIRs). Road sign classification is based on the description of RGB components and color oriented elements. The correct determination of three facial poses by the cascade method in the automatic detection of the MIRs [1] was 0.87. The recognition rate for the actual road signs using the color and shape classification algorithm was 96%.

[Full paper]



Back to Publication List
Лист публикаций