The method of false vortices in designation problems of targets, road signs, facial features and real vortices in space images

  • Evgeni Nikolaevich Terentiev1 1 Department of Computer Modeling and Informatics, Faculty of Physics, M.V.Lomonosov Moscow State University, Moscow, Russia
  • I. García-Magariño2 2 University of Zaragoza, Zaragoza, Aragon, Spain
  • N. E. Shilin-Terentyev3 3 Google, San Francisco, CA, USA
  • I. N. Prikhodko1
  • I. I. Farshakova1
Keywords: Image Analysis, Localization of Object in the Image, Computer vision, Designation Problems of Targets, Road Signs, Facial Features, False Vortices

Abstract

An image in one color is a scalar field. The image is associated with its vector gradient field. The analysis of vector fields with the help of single-length vector templates made it possible to indicate in the image such objects as vortices, Road Signs (RS), Facial Features (FF), and so on. The proposed method of indicating local objects using gradient fields turned out to be highly accurate and noise-proof. The presented advantages of the proposed method recommend it as a promising tool in targets designation problems.

Author Biography

Evgeni Nikolaevich Terentiev1, 1 Department of Computer Modeling and Informatics, Faculty of Physics, M.V.Lomonosov Moscow State University, Moscow, Russia

Terentiev E.N. received Ph.D in the mathematical physics in 1982.
The areas of interest are the models of image formation through turbulent and scattering atmosphere in a microwave and an optical wave range, the direct and inverse (ill-posed) problems, the laser beam focusing in turbulent atmosphere, computer methods for physics and interpretations of models.
He works for Mathematical Modeling and Informatics Department of Physics Faculty, M.V. Lomonosov Moscow State University.
He is author of 34 papers in Pros. SPIE.

Published
2018-12-03
Section
Articles