Image-based identification of observed aerodynamical object

Аuthors

Samoilenko M. V.

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

e-mail: samoi.mar@mail.ru

Abstract

An identification problem is solved for some observed aerodynamical object based on its lengthy image. Collections of reference object images observed for various angles are used as an a priori information for comparison purposes. A technique is offered to choose reference images from the collections. The technique allows to pick out some reference image with the same angle as observed one. A using of binary images is proved to identify observed object. It is substantiated also an utilization of decision making algorithm based on maximum coincidence of identified and reference image shapes. A special index is proposed to evaluate the coincidence for image shapes. A considerable advantage is shown for this index in comparison with frequently used correlation one.