Infrared Target Detection Using Intensity Saliency and Self-Attention

Zhang, R; Xu, M; Shi, Y; Fan, J; Mu, C; Xu, L

Infrared Target Detection Using Intensity Saliency and Self-Attention

Zhang, R

Xu, M

Shi, Y Fan, J Mu, C Xu, L

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: Proceedings - International Conference on Image Processing, ICIP, 2020, 2020-October, pp. 1991-1995
Issue Date:: 2020-10-01

Closed Access

	Filename	Description	Size
	09191055.pdf	Published version	257.76 kB	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	Zhang, R https://orcid.org/0000-0002-5460-7196
dc.contributor.author	Xu, M https://orcid.org/0000-0001-9581-8849
dc.contributor.author	Shi, Y
dc.contributor.author	Fan, J
dc.contributor.author	Mu, C
dc.contributor.author	Xu, L
dc.date	2020-10-25
dc.date.accessioned	2021-05-27T07:11:32Z
dc.date.available	2021-05-27T07:11:32Z
dc.date.issued	2020-10-01
dc.identifier.citation	Proceedings - International Conference on Image Processing, ICIP, 2020, 2020-October, pp. 1991-1995
dc.identifier.isbn	9781728163956
dc.identifier.issn	1522-4880
dc.identifier.uri	http://hdl.handle.net/10453/149268
dc.description.abstract	Infrared target detection is essential for many computer vision tasks. Generally, the IR images present common infrared characteristics, such as poor texture information, low resolution, and high noise. However, these characteristics are ignored in the existing detection methods, making them fail in real-world scenarios. In this paper, we take infrared intensity into account and propose a novel backbone network named Deep-IRTarget. We first extract infrared intensity saliency by a convolution with a Gaussian kernel filtering the images in the frequency domain. We then propose the triple self-attention network to further extract spatial domain image saliency by selectively emphasize interdependent semantic features in each channel. Jointly exploiting infrared characteristics in the frequency domain and the overall semantic interdependencies in the spatial domain, the proposed Deep-IRTarget outperforms existing methods in real-world Infrared target detection tasks. Experimental results on two infrared imagery datasets demonstrate the superiorly of our model.
dc.language	en
dc.publisher	IEEE
dc.relation.ispartof	Proceedings - International Conference on Image Processing, ICIP
dc.relation.ispartof	IEEE International Conference on Image Processing
dc.relation.isbasedon	10.1109/ICIP40778.2020.9191055
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Infrared Target Detection Using Intensity Saliency and Self-Attention
dc.type	Conference Proceeding
utslib.citation.volume	2020-October
utslib.location.activity	Abu Dhabi, United Arab
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/Strength - INEXT - Innovation in IT Services and Applications
pubs.organisational-group	/University of Technology Sydney/Strength - GBDTC - Global Big Data Technologies
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/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	closed_access	*
pubs.consider-herdc	false
dc.date.updated	2021-05-27T07:11:31Z
pubs.finish-date	2020-10-28
pubs.publication-status	Published
pubs.start-date	2020-10-25
pubs.volume	2020-October

Abstract:

Infrared target detection is essential for many computer vision tasks. Generally, the IR images present common infrared characteristics, such as poor texture information, low resolution, and high noise. However, these characteristics are ignored in the existing detection methods, making them fail in real-world scenarios. In this paper, we take infrared intensity into account and propose a novel backbone network named Deep-IRTarget. We first extract infrared intensity saliency by a convolution with a Gaussian kernel filtering the images in the frequency domain. We then propose the triple self-attention network to further extract spatial domain image saliency by selectively emphasize interdependent semantic features in each channel. Jointly exploiting infrared characteristics in the frequency domain and the overall semantic interdependencies in the spatial domain, the proposed Deep-IRTarget outperforms existing methods in real-world Infrared target detection tasks. Experimental results on two infrared imagery datasets demonstrate the superiorly of our model.

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

http://hdl.handle.net/10453/149268