TELECOMMUNICATIONS AND RADIO ENGINEERING - 2011 Vol. 70,
No 15 		 

 

 

 

COMPRESSION OF DESCRIPTIONS IN THE STRUCTURAL IMAGE RECOGNITION



V.A. Gorokhovatskiy
Kharkov National University of Radio Engineering and Electronics,
14, Lenin Ave, Kharkiv, 61166, Ukraine
E-mail: gorohovatsky-v@rambler.ru

Abstract
Approaches to the gain in efficiency of the structural-hierarchical methods for image recognition using the principle of voting are considered. Results of elaboration of the modified methods, which through the structural description compression attain significantly lower calculation expenses as compared to the traditional approaches while ensuring a high probability of correct classification, are offered. Results of experiments in the images analysis and classification, based on the offered methods, are considered.
KEY WORDS:images recognition and classification, structural-hierarchical methods, structural description, principle of voting, compression of description, calculation expenses, probability of correct classification

References

  1. Shapiro, L., (2001), Computer vision, Prentice Hall, - 625 p.
  2. Shlezinger, M. and Glavach, V., (2004), Ten lectures on statistical and structural recognition, Naukova Dumka, Kiev: 535 p. (in Russian).
  3. Gorokhovatsky, V.À. and Putyatin, Ye.P., (2008), Structural recognition of images on the basis of voting models of attributes of typical points, Data recording, storage and processing. 10(4):75-85 (in Russian).
  4. Gorokhovatsky, V.À., (2010), Local-spatial similarity modeles of structural description of virtue objects, Systems of control, navigation and communications. 3(15):82–85(in Russian).
  5. Gorokhovatsky, V.A. and Putyatin, Ye.P., (2009), Image Likelihood Measures of the Basis of the Set of Conformities, Telecommunications and Radio Engineering. 68(9):763-778.
  6. Gorokhovatsky, V.À. and Polyakova, T.V., (2009), Increasing in speed of method of afinno-invariants in structural description of visual objects, Systems of control, navigation and communications. 2(10):81–84(in Russian).
  7. Àyvazyan, S.À., Buchstaber, V.Ì., Yenyukov, I.S., and Ìåshalkin, L.D., (1989), Appliedstatistics: Classificationanddecreasingof dimensions. Finance and statistics, Moscow: 607 p. (in Russian).
  8. Gorokhovatsky, V.À. and Grebennik, P.V., (2009), Hierarchical measures in structural recognition of images with description compression, Systems of control, navigation and communications. 11(3):26–33 (in Russian).
  9. Gorokhovatsky, V.À., (2008), Hierarchy of spatial interrelations between structural attributes in the problems of comparison of visual objects, Systems of control, navigation and communication. 3(7):85-89 (in Russian).
  10. Sytnik, O.V. and Gorohovatsky, A.V., (2007), Signal processing algorithms in identification of subsurface objects, Radioelectronics and Communications Systems, 50(10):557-563 (in Russian).
  11. Rossi, F., van Beek P., and Walsh, T., (eds.) (2006), Handbook of Constraint Programming. Foundations of Artificial Intelligence. Elsevier edition, - 955p.


pages 1363-1371

Back