Bottom-up visual saliency map using wavelet transform domain

Christopher Wing Hong Ngau, Li Minn Ang, Kah Phooi Seng

Research output: Book chapter/Published conference paperConference paper

13 Citations (Scopus)

Abstract

Object or region based image processing can be performed more efficiently with information pertaining locations that are visually salient to human perception with the aid of a saliency map. The saliency map is a master topological map having the possible locations of objects or regions which a human perceived as important/salient. In this paper, a method to compute the saliency map in the wavelet transform domain is explored. Previous works involving saliency in this domain usually involves salient points, which are in fact accurate but they do not cover the area as a region and involve heavy repeated iterations. The method explored in this paper is compared to two state-of-art methods in which these methods involve the frequency domain. The presented method provides more accurate salient regions compared to the other two methods while retaining a resolution which the salient regions are visually identifiable.

Original languageEnglish
Title of host publicationProceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010
Pages692-695
Number of pages4
Volume1
DOIs
Publication statusPublished - 01 Nov 2010
Event2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010 - Chengdu, China
Duration: 09 Jul 201011 Jul 2010

Conference

Conference2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010
CountryChina
CityChengdu
Period09/07/1011/07/10

Fingerprint

Wavelet transforms
Image processing

Cite this

Ngau, C. W. H., Ang, L. M., & Seng, K. P. (2010). Bottom-up visual saliency map using wavelet transform domain. In Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010 (Vol. 1, pp. 692-695). [5564545] https://doi.org/10.1109/ICCSIT.2010.5564545
Ngau, Christopher Wing Hong ; Ang, Li Minn ; Seng, Kah Phooi. / Bottom-up visual saliency map using wavelet transform domain. Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010. Vol. 1 2010. pp. 692-695
@inproceedings{033cc227f8f2411ab1e419689dc7ae82,
title = "Bottom-up visual saliency map using wavelet transform domain",
abstract = "Object or region based image processing can be performed more efficiently with information pertaining locations that are visually salient to human perception with the aid of a saliency map. The saliency map is a master topological map having the possible locations of objects or regions which a human perceived as important/salient. In this paper, a method to compute the saliency map in the wavelet transform domain is explored. Previous works involving saliency in this domain usually involves salient points, which are in fact accurate but they do not cover the area as a region and involve heavy repeated iterations. The method explored in this paper is compared to two state-of-art methods in which these methods involve the frequency domain. The presented method provides more accurate salient regions compared to the other two methods while retaining a resolution which the salient regions are visually identifiable.",
keywords = "Frequency-time domain, Saliency map, Salient region, Wavelet transform",
author = "Ngau, {Christopher Wing Hong} and Ang, {Li Minn} and Seng, {Kah Phooi}",
year = "2010",
month = "11",
day = "1",
doi = "10.1109/ICCSIT.2010.5564545",
language = "English",
isbn = "9781424455386",
volume = "1",
pages = "692--695",
booktitle = "Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010",

}

Ngau, CWH, Ang, LM & Seng, KP 2010, Bottom-up visual saliency map using wavelet transform domain. in Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010. vol. 1, 5564545, pp. 692-695, 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010, Chengdu, China, 09/07/10. https://doi.org/10.1109/ICCSIT.2010.5564545

Bottom-up visual saliency map using wavelet transform domain. / Ngau, Christopher Wing Hong; Ang, Li Minn; Seng, Kah Phooi.

Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010. Vol. 1 2010. p. 692-695 5564545.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Bottom-up visual saliency map using wavelet transform domain

AU - Ngau, Christopher Wing Hong

AU - Ang, Li Minn

AU - Seng, Kah Phooi

PY - 2010/11/1

Y1 - 2010/11/1

N2 - Object or region based image processing can be performed more efficiently with information pertaining locations that are visually salient to human perception with the aid of a saliency map. The saliency map is a master topological map having the possible locations of objects or regions which a human perceived as important/salient. In this paper, a method to compute the saliency map in the wavelet transform domain is explored. Previous works involving saliency in this domain usually involves salient points, which are in fact accurate but they do not cover the area as a region and involve heavy repeated iterations. The method explored in this paper is compared to two state-of-art methods in which these methods involve the frequency domain. The presented method provides more accurate salient regions compared to the other two methods while retaining a resolution which the salient regions are visually identifiable.

AB - Object or region based image processing can be performed more efficiently with information pertaining locations that are visually salient to human perception with the aid of a saliency map. The saliency map is a master topological map having the possible locations of objects or regions which a human perceived as important/salient. In this paper, a method to compute the saliency map in the wavelet transform domain is explored. Previous works involving saliency in this domain usually involves salient points, which are in fact accurate but they do not cover the area as a region and involve heavy repeated iterations. The method explored in this paper is compared to two state-of-art methods in which these methods involve the frequency domain. The presented method provides more accurate salient regions compared to the other two methods while retaining a resolution which the salient regions are visually identifiable.

KW - Frequency-time domain

KW - Saliency map

KW - Salient region

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=77958601297&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77958601297&partnerID=8YFLogxK

U2 - 10.1109/ICCSIT.2010.5564545

DO - 10.1109/ICCSIT.2010.5564545

M3 - Conference paper

AN - SCOPUS:77958601297

SN - 9781424455386

VL - 1

SP - 692

EP - 695

BT - Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010

ER -

Ngau CWH, Ang LM, Seng KP. Bottom-up visual saliency map using wavelet transform domain. In Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010. Vol. 1. 2010. p. 692-695. 5564545 https://doi.org/10.1109/ICCSIT.2010.5564545