Analysis of object arrangement by means of neural networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F44555601%3A13430%2F01%3A00001408" target="_blank" >RIV/44555601:13430/01:00001408 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Analysis of object arrangement by means of neural networks

Popis výsledku v původním jazyce

In this contribution a new approach to the image analysis of the systems consisting of a large number of individual objects based on neural networks is presented. This technique can profoundly simplify the image analysis of more complicated systems. Thecontribution presents and compares some morphological methods as pre-processing algorithms for neural networks, which is a principal task for a neural networks application. It has been shown that the use of the networks for the determination of the pointstructure randomness leads to successful results. The new approach was tested on simulated hard-disk models.The quality of the performance of the trained networks for real images is a subject of further investigation, now

Název v anglickém jazyce

Analysis of object arrangement by means of neural networks

Popis výsledku anglicky

In this contribution a new approach to the image analysis of the systems consisting of a large number of individual objects based on neural networks is presented. This technique can profoundly simplify the image analysis of more complicated systems. Thecontribution presents and compares some morphological methods as pre-processing algorithms for neural networks, which is a principal task for a neural networks application. It has been shown that the use of the networks for the determination of the pointstructure randomness leads to successful results. The new approach was tested on simulated hard-disk models.The quality of the performance of the trained networks for real images is a subject of further investigation, now

Druh

J_x - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)

CEP obor

BL - Fyzika plasmatu a výboje v plynech

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2001

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Ergo (Ústí nad Labem)

ISSN

1212-8317

e-ISSN

—

Svazek periodika

2001

Číslo periodika v rámci svazku

3

Stát vydavatele periodika

CZ - Česká republika

Počet stran výsledku

14

Strana od-do

83-96

Kód UT WoS článku

—

EID výsledku v databázi Scopus

—