Structural image classification with graph neural networks

Quek, A; Wang, Z; Zhang, J; Feng, D

Structural image classification with graph neural networks

Quek, A Wang, Z Zhang, J

Feng, D

Permalink

Publication Type:: Conference Proceeding
Citation:: Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011, 2011, pp. 416 - 421
Issue Date:: 2011-12-01

Closed Access

	Filename	Description	Size
	2011001961OK.pdf		2.74 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Quek, A	en_US
dc.contributor.author	Wang, Z	en_US
dc.contributor.author	Zhang, J https://orcid.org/0000-0002-7240-3541	en_US
dc.contributor.author	Feng, D	en_US
dc.date.issued	2011-12-01	en_US
dc.identifier.citation	Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011, 2011, pp. 416 - 421	en_US
dc.identifier.isbn	9780769545882	en_US
dc.identifier.uri	http://hdl.handle.net/10453/19153
dc.description.abstract	Many approaches to image classification tend to transform an image into an unstructured set of numeric feature vectors obtained globally and/or locally, and as a result lose important relational information between regions. In order to encode the geometric relationships between image regions, we propose a variety of structural image representations that are not specialised for any particular image category. Besides the traditional grid-partitioning and global segmentation methods, we investigate the use of local scale-invariant region detectors. Regions are connected based not only upon nearest-neighbour heuristics, but also upon minimum spanning trees and Delaunay triangulation. In order to maintain the topological and spatial relationships between regions, and also to effectively process undirected connections represented as graphs, we utilise the recently-proposed graph neural network model. To the best of our knowledge, this is the first utilisation of the model to process graph structures based on local-sampling techniques, for the task of image classification. Our experimental results demonstrate great potential for further work in this domain. © 2011 IEEE.	en_US
dc.relation.ispartof	Proceedings - 2011 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011	en_US
dc.relation.isbasedon	10.1109/DICTA.2011.77	en_US
dc.title	Structural image classification with graph neural networks	en_US
dc.type	Conference Proceeding
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
dc.location.activity	Noosa, Queensland, Australia	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
pubs.organisational-group	/University of Technology Sydney/Strength - GBDTC - Global Big Data Technologies
utslib.copyright.status	closed_access
pubs.publication-status	Published	en_US

Abstract:

Many approaches to image classification tend to transform an image into an unstructured set of numeric feature vectors obtained globally and/or locally, and as a result lose important relational information between regions. In order to encode the geometric relationships between image regions, we propose a variety of structural image representations that are not specialised for any particular image category. Besides the traditional grid-partitioning and global segmentation methods, we investigate the use of local scale-invariant region detectors. Regions are connected based not only upon nearest-neighbour heuristics, but also upon minimum spanning trees and Delaunay triangulation. In order to maintain the topological and spatial relationships between regions, and also to effectively process undirected connections represented as graphs, we utilise the recently-proposed graph neural network model. To the best of our knowledge, this is the first utilisation of the model to process graph structures based on local-sampling techniques, for the task of image classification. Our experimental results demonstrate great potential for further work in this domain. © 2011 IEEE.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/19153