We present a simple and general-purpose method able to combine high-resolution procedure with the classification and identification of objects of interest from microscopy imaging. The method is composed of two stages. First (pattern recognition), promising components (possible objects of interest) in the image are detected and small regions containing the objects of interest are extracted using a feature finder. Second, high-resolution algorithms are applied to such identified components in order to approach a multiple scales of resolution. Although the method is indeed to be applied to any microscopy technique, in this paper, we have focused the attention on biological systems, like animal cells, recorded with an atomic force microscopy.
A new method combining enhanced resolution and pattern identification
D'Acunto M;Righi M;Salvetti O
2016
Abstract
We present a simple and general-purpose method able to combine high-resolution procedure with the classification and identification of objects of interest from microscopy imaging. The method is composed of two stages. First (pattern recognition), promising components (possible objects of interest) in the image are detected and small regions containing the objects of interest are extracted using a feature finder. Second, high-resolution algorithms are applied to such identified components in order to approach a multiple scales of resolution. Although the method is indeed to be applied to any microscopy technique, in this paper, we have focused the attention on biological systems, like animal cells, recorded with an atomic force microscopy.| File | Dimensione | Formato | |
|---|---|---|---|
|
prod_357485-doc_116928.pdf
solo utenti autorizzati
Descrizione: A new method combining enhanced resolution and pattern identification
Tipologia:
Versione Editoriale (PDF)
Dimensione
782.91 kB
Formato
Adobe PDF
|
782.91 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
|
prod_357485-doc_166328.pdf
accesso aperto
Descrizione: Postprint - A new method combining enhanced resolution and pattern identification
Tipologia:
Versione Editoriale (PDF)
Dimensione
401.23 kB
Formato
Adobe PDF
|
401.23 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


