EFFICIENCY OF THE METHOD OF COMPENSATION OF INFORMATION FLOWS OF THE REFERENCE IMAGE GENERATION SYSTEM
Background. One of the main ways to improve the efficiency of information flow compensation is to build a recognition system for the method of information flow compensation for generating reference images. Materials and methods. If a class of systems is characterized by a list of its members, it c...
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | article |
| Lenguaje: | EN RU |
| Publicado: |
Penza State University Publishing House
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/34b5d1284be94060a6f20783876100bc |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Background. One of the main ways to improve the efficiency of information flow compensation is to build
a recognition system for the method of information flow compensation for generating reference images. Materials
and methods. If a class of systems is characterized by a list of its members, it can be based on the principle of belonging to
this list – this method is called the method of comparison with the standard. According to this method, many images belonging
to the same class are stored by the identification system. When new images are presented to the system, it consistently
compares them with the reference images stored in memory. The system assigns the new image to the class that
the reference image that coincided with the presented one belonged to. Results and conclusions. This method works well only
when the sample is close to ideal, or the conditions for presenting images are ideal. If members of the same class have
some common features, the recognition system is based on the principle of common properties. These General properties
are stored in the system memory. During image processing, the system must be able to extract features from the presented
image and work with them. The system assigns the newly presented image to a class whose features are similar to those
highlighted in the new image. |
|---|