(24) * << * >> * Russian * English * Content * All Issues

Searching for cavities in a closed non-convex contour

R.V. Khmelev1,2
1 Image Processing Systems Institute of RAS;

 2 Samara State Aerospace University

 PDF, 522 kB

Pages: 164 - 172.

Abstract:
It is known that structural methods of object recognition are based on the search for elementary object components and analysis of their relative position [3]. However, this search for elementary components is complicated, partly because of the complexity of formalization, and partly because of their diversity. In order to create recognition methods comparable to a human being in terms of reliability, it is probably necessary to formalize all the concepts, properties and definitions of Euclidean geometry, since all of these were once noticed and identified by a person.

Keywords:
non-convex contour, structural method of object, recognition method, Euclidean geometry.


Citation:
Khmelev RV. Searching for cavities in a closed non-convex contour. Computer Optics 2002; 24: 164-172.

References:
  1. Mikhlyaev SV. Relaxation technique for stereo images matching. Proc ACIT 2002: 376-381.
  2. Chetverikov D, Szabó Z. Detection of high curvature points in planar curves. Source: http://visual.ipan.sztaki.hu/corner/.
  3. Tou JT, Gonzalez RC. Pattern recognition principles. 2nd ed. Addison-Wesley; 1977.
  4. Furman YaA, ed. Introduction to contour analysis and its applications to image and signal processing [In Russian]. Moscow: "Fizmatlit" Publisher; 2002.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20