Multi-resolution analysis based salient contour extraction

Jing Wang, Kazuo Kunieda, Makoto Iwata, Hirokazu Koizumi, Hideo Shimazu, Takeshi Ikenaga, Satoshi Goto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Detecting salient contours in complex backgrounds is important in image analysis and scene understanding. The local context of an edge or line segment feature is commonly used to measure its saliency degree as a part of the object boundary. However, traditionally the context information is captured by studying several features in the predefined neighborhood. In this paper, a novel salient contour extraction algorithm based on the multi-resolution analysis is proposed and a new saliency measure is defined to characterize the significance of feature. Relation of features that are corresponding to the same part of the object boundary across resolutions is utilized to estimate the context information and feature significance value. Experimental results show that the proposed method can extract salient contours more efficiently than center-surround interaction based methods and still provide robust results.

Original languageEnglish
Title of host publication2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06
Pages689-692
Number of pages4
DOIs
Publication statusPublished - 2007
Event2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06 - Yonago
Duration: 2006 Dec 122006 Dec 15

Other

Other2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06
CityYonago
Period06/12/1206/12/15

Fingerprint

Contour Extraction
Multiresolution analysis
Multiresolution Analysis
Image analysis
Saliency
Line segment
Image Analysis
Experimental Results
Interaction
Estimate
Context
Object

ASJC Scopus subject areas

  • Computer Science Applications
  • Signal Processing
  • Software
  • Electrical and Electronic Engineering
  • Mathematics(all)

Cite this

Wang, J., Kunieda, K., Iwata, M., Koizumi, H., Shimazu, H., Ikenaga, T., & Goto, S. (2007). Multi-resolution analysis based salient contour extraction. In 2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06 (pp. 689-692). [4212367] https://doi.org/10.1109/ISPACS.2006.364749

Multi-resolution analysis based salient contour extraction. / Wang, Jing; Kunieda, Kazuo; Iwata, Makoto; Koizumi, Hirokazu; Shimazu, Hideo; Ikenaga, Takeshi; Goto, Satoshi.

2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06. 2007. p. 689-692 4212367.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, J, Kunieda, K, Iwata, M, Koizumi, H, Shimazu, H, Ikenaga, T & Goto, S 2007, Multi-resolution analysis based salient contour extraction. in 2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06., 4212367, pp. 689-692, 2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06, Yonago, 06/12/12. https://doi.org/10.1109/ISPACS.2006.364749
Wang J, Kunieda K, Iwata M, Koizumi H, Shimazu H, Ikenaga T et al. Multi-resolution analysis based salient contour extraction. In 2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06. 2007. p. 689-692. 4212367 https://doi.org/10.1109/ISPACS.2006.364749
Wang, Jing ; Kunieda, Kazuo ; Iwata, Makoto ; Koizumi, Hirokazu ; Shimazu, Hideo ; Ikenaga, Takeshi ; Goto, Satoshi. / Multi-resolution analysis based salient contour extraction. 2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06. 2007. pp. 689-692
@inproceedings{92e4ce47b62243deb01bcf61f4717abe,
title = "Multi-resolution analysis based salient contour extraction",
abstract = "Detecting salient contours in complex backgrounds is important in image analysis and scene understanding. The local context of an edge or line segment feature is commonly used to measure its saliency degree as a part of the object boundary. However, traditionally the context information is captured by studying several features in the predefined neighborhood. In this paper, a novel salient contour extraction algorithm based on the multi-resolution analysis is proposed and a new saliency measure is defined to characterize the significance of feature. Relation of features that are corresponding to the same part of the object boundary across resolutions is utilized to estimate the context information and feature significance value. Experimental results show that the proposed method can extract salient contours more efficiently than center-surround interaction based methods and still provide robust results.",
author = "Jing Wang and Kazuo Kunieda and Makoto Iwata and Hirokazu Koizumi and Hideo Shimazu and Takeshi Ikenaga and Satoshi Goto",
year = "2007",
doi = "10.1109/ISPACS.2006.364749",
language = "English",
isbn = "0780397339",
pages = "689--692",
booktitle = "2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06",

}

TY - GEN

T1 - Multi-resolution analysis based salient contour extraction

AU - Wang, Jing

AU - Kunieda, Kazuo

AU - Iwata, Makoto

AU - Koizumi, Hirokazu

AU - Shimazu, Hideo

AU - Ikenaga, Takeshi

AU - Goto, Satoshi

PY - 2007

Y1 - 2007

N2 - Detecting salient contours in complex backgrounds is important in image analysis and scene understanding. The local context of an edge or line segment feature is commonly used to measure its saliency degree as a part of the object boundary. However, traditionally the context information is captured by studying several features in the predefined neighborhood. In this paper, a novel salient contour extraction algorithm based on the multi-resolution analysis is proposed and a new saliency measure is defined to characterize the significance of feature. Relation of features that are corresponding to the same part of the object boundary across resolutions is utilized to estimate the context information and feature significance value. Experimental results show that the proposed method can extract salient contours more efficiently than center-surround interaction based methods and still provide robust results.

AB - Detecting salient contours in complex backgrounds is important in image analysis and scene understanding. The local context of an edge or line segment feature is commonly used to measure its saliency degree as a part of the object boundary. However, traditionally the context information is captured by studying several features in the predefined neighborhood. In this paper, a novel salient contour extraction algorithm based on the multi-resolution analysis is proposed and a new saliency measure is defined to characterize the significance of feature. Relation of features that are corresponding to the same part of the object boundary across resolutions is utilized to estimate the context information and feature significance value. Experimental results show that the proposed method can extract salient contours more efficiently than center-surround interaction based methods and still provide robust results.

UR - http://www.scopus.com/inward/record.url?scp=45249087078&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=45249087078&partnerID=8YFLogxK

U2 - 10.1109/ISPACS.2006.364749

DO - 10.1109/ISPACS.2006.364749

M3 - Conference contribution

SN - 0780397339

SN - 9780780397330

SP - 689

EP - 692

BT - 2006 International Symposium on Intelligent Signal Processing and Communications, ISPACS'06

ER -