Structural Resonance Methods for Image Processing and Pattern Recognition
| Parent link: | International Journal of Applied Engineering Research: International Research Journal Vol. 12, № 19.— 2017.— [P. 9087-9088] |
|---|---|
| Corporate Author: | |
| Other Authors: | , , , , , , , |
| Summary: | Title screen We have developed group-theoretical methods of structuralresonance by analogy with "resonance methods" applied inthe classical theory of oscillations and waves and in quantumphysics. The "structural-resonance approach" is one of theadvantageous interpretations of the reconstructivecomputerized diagnostics based on the previously developedgroup-theoretical statistical approach to solve ill-posedinverse problems. We have elaborated a unified method ofreconstructing a “heterogenous” test-object with widesemantic spectrum, regarding this object as a “semanticmixture” of different mutually complementary semanticcontents, matching each of them with its formalized invariantstructure. To separate a cleared semantic content from the“mixture”, it is necessary to apply operators from its group ofautomorphisms to the informational image of the “mixture”. Itresults in randomization of all other semantic contents, so theycan be easily suppressed statistically. We have definedmeasures of intensity for such resonance, carried out adetailed comparison of tomosynthesis and such structuralresonance “sense synthesis”, and compared “materialreconstruction” and reconstruction of structural-functionalconnections. Finally yet importantly, we have discussedprospects of these methods for solving ill-posed problems invarious fields of science and practice. Режим доступа: по договору с организацией-держателем ресурса |
| Published: |
2017
|
| Subjects: | |
| Online Access: | https://www.ripublication.com/ijaer17/ijaerv12n19_145.pdf |
| Format: | Electronic Book Chapter |
| KOHA link: | https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=660154 |