Graph-based segmentation with homogeneous hue and texture vertices - Lower Silesian Digital Library

Skip to main menu
Skip to search engine
Skip to content
Skip to footer

Planned object

Title: Graph-based segmentation with homogeneous hue and texture vertices

Creator:

Ngo, Lua ; Han, Jae-Ho

Contributor:

Urbańczyk, Wacław. Redakcja

Description:

Optica Applicata, Vol. 51, 2021, nr 4, s. 541-549 ; Optica Applicata is an international journal, published in a non-periodical form in the years 1971-1973 and quarterly since 1973. From the beginning of the year 2008, Optica Applicata is an Open Access journal available online via the Internet, with free access to the full text of articles serving the best interests of the scientific community. The journal is abstracted and indexed in: Chemical Abstracts, Compendex, Current Contents, Inspec, Referativnyj Zhurnal, SCI Expanded, Scopus, Ulrich’s Periodicals Directory ; click here to follow the link

Abstrakt:

This work presents an automated segmentation method, based on graph theory, which processes superpixels that exhibit spatially similarities in hue and texture pixel groups, rather than individual pixels. The graph shortest path includes a chain of neighboring superpixels which have minimal intensity changes. This method reduces graphics computational complexity because it provides large decreases in the number of vertices as the superpixel size increases. For the starting vertex prediction, the boundary pixel in first column which is included in this starting vertex is predicted by a trained deep neural network formulated as a regression task. By formulating the problem as a regression scheme, the computational burden is decreased in comparison with classifying each pixel in the entire image. This feasibility approach, when applied as a preliminary study in electron microscopy and optical coherence tomography images, demonstrated high measures of accuracy: 0.9670 for the electron microscopy image and 0.9930 for vitreous/nerve-fiber and inner-segment/outer-segment layer segmentations in the optical coherence tomography image.

Publisher:

Oficyna Wydawnicza Politechniki Wrocławskiej

Place of publication:

Date:

Resource Type:

Resource Identifier:

doi:10.37190/oa210406

Source:

<sygn. PWr A3481II> ; click here to follow the link ; click here to follow the link

Language:

Relation:

Optica Applicata ; Optica Applicata, Vol. 51, 2021 ; Optica Applicata, Vol. 51, 2021, nr 4 ; Politechnika Wrocławska. Wydział Podstawowych Problemów Techniki

Rights:

Wszystkie prawa zastrzeżone (Copyright)

Access Rights:

Dla wszystkich w zakresie dozwolonego użytku

Location:

Politechnika Wrocławska

Group publication title:

Optica Applicata

Subject and keywords:

optyka image segmentation deep neural network electron microscopy optical coherence tomography (OCT) pattern recognition

Object collections:

Last modified:

Feb 7, 2025

All available object's versions:

https://dbc.wroc.pl./publication/169938

Show description in RDF format:

Optica Applicata

Edition name	Date

Citation style:

Chicago ISO690 APA IEEE Harvard

Contact details

Address

Wrocław University of Science and Technology (Coordinator)
Library of Wrocław University of Science and Technology
Wybrzeże Wyspiańskiego 27
50-370 Wrocław, Poland

Phone

(+48) 71-320-38-30
(+48) 71-320-47-50

E-Mail

Visit us!

https://biblioteka.pwr.edu.pl/

This page uses 'cookies'. More information