Texture image segmentation using spectral clustering

Hui Du; Yuping Wang; Xiaopan Dong; Yiu Ming CHEUNG

doi:10.1007/978-3-319-21380-4_113

Texture image segmentation using spectral clustering

Hui Du, Yuping Wang^*, Xiaopan Dong, Yiu Ming CHEUNG

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

3 Citations (Scopus)

Abstract

Clustering is a popular and effective method for texture image segmentation. However, most cluster methods often suffer the following problems: need a huge space and a lot of computation when the input data is large. To save the space and computation, we construct a novel algorithm for image segmentation. It consists of two phases: Sampling and clustering. First, we put some detectors into the data space uniformly using orthogonal design method. These detectors can move and merge according to the law of universal gravitation. When the detectors are in a stable status (i.e., do not move), these detectors are used as the representative samples to the next step. Second, to further improve the efficiency and avoid dependence on parameters, the Self-tuning Spectral Clustering (SSC) is used to the representative samples to do the clustering. As a result, the proposed algorithm can quickly and precisely realize the clustering for texture image segmentation.

Original language	English
Title of host publication	HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings
Editors	Constantine Stephanidis
Publisher	Springer Verlag
Pages	671-676
Number of pages	6
ISBN (Print)	9783319213798
DOIs	https://doi.org/10.1007/978-3-319-21380-4_113
Publication status	Published - 2015
Event	17th International Conference on Human Computer Interaction, HCI 2015 - Los Angeles, United States Duration: 2 Aug 2015 → 7 Aug 2015

Publication series

Name	Communications in Computer and Information Science
Volume	528
ISSN (Print)	1865-0929

Conference

Conference	17th International Conference on Human Computer Interaction, HCI 2015
Country/Territory	United States
City	Los Angeles
Period	2/08/15 → 7/08/15

Scopus Subject Areas

Computer Science(all)
Mathematics(all)

User-Defined Keywords

Feature extraction
Gray level histogram
Sampling
Spectral clustering
Texture image segmentation

Access to Document

10.1007/978-3-319-21380-4_113

Cite this

Du, H., Wang, Y., Dong, X., & CHEUNG, Y. M. (2015). Texture image segmentation using spectral clustering. In C. Stephanidis (Ed.), HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings (pp. 671-676). (Communications in Computer and Information Science; Vol. 528). Springer Verlag. https://doi.org/10.1007/978-3-319-21380-4_113

@inproceedings{6d6a79d3f596474b9f94d1c8516f2e0c,

title = "Texture image segmentation using spectral clustering",

abstract = "Clustering is a popular and effective method for texture image segmentation. However, most cluster methods often suffer the following problems: need a huge space and a lot of computation when the input data is large. To save the space and computation, we construct a novel algorithm for image segmentation. It consists of two phases: Sampling and clustering. First, we put some detectors into the data space uniformly using orthogonal design method. These detectors can move and merge according to the law of universal gravitation. When the detectors are in a stable status (i.e., do not move), these detectors are used as the representative samples to the next step. Second, to further improve the efficiency and avoid dependence on parameters, the Self-tuning Spectral Clustering (SSC) is used to the representative samples to do the clustering. As a result, the proposed algorithm can quickly and precisely realize the clustering for texture image segmentation.",

keywords = "Feature extraction, Gray level histogram, Sampling, Spectral clustering, Texture image segmentation",

author = "Hui Du and Yuping Wang and Xiaopan Dong and CHEUNG, {Yiu Ming}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.; 17th International Conference on Human Computer Interaction, HCI 2015 ; Conference date: 02-08-2015 Through 07-08-2015",

year = "2015",

doi = "10.1007/978-3-319-21380-4_113",

language = "English",

isbn = "9783319213798",

series = "Communications in Computer and Information Science",

publisher = "Springer Verlag",

pages = "671--676",

editor = "Constantine Stephanidis",

booktitle = "HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings",

address = "Germany",

}

Du, H, Wang, Y, Dong, X & CHEUNG, YM 2015, Texture image segmentation using spectral clustering. in C Stephanidis (ed.), HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings. Communications in Computer and Information Science, vol. 528, Springer Verlag, pp. 671-676, 17th International Conference on Human Computer Interaction, HCI 2015, Los Angeles, United States, 2/08/15. https://doi.org/10.1007/978-3-319-21380-4_113

Texture image segmentation using spectral clustering. / Du, Hui; Wang, Yuping; Dong, Xiaopan et al.
HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings. ed. / Constantine Stephanidis. Springer Verlag, 2015. p. 671-676 (Communications in Computer and Information Science; Vol. 528).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Texture image segmentation using spectral clustering

AU - Du, Hui

AU - Wang, Yuping

AU - Dong, Xiaopan

AU - CHEUNG, Yiu Ming

PY - 2015

Y1 - 2015

N2 - Clustering is a popular and effective method for texture image segmentation. However, most cluster methods often suffer the following problems: need a huge space and a lot of computation when the input data is large. To save the space and computation, we construct a novel algorithm for image segmentation. It consists of two phases: Sampling and clustering. First, we put some detectors into the data space uniformly using orthogonal design method. These detectors can move and merge according to the law of universal gravitation. When the detectors are in a stable status (i.e., do not move), these detectors are used as the representative samples to the next step. Second, to further improve the efficiency and avoid dependence on parameters, the Self-tuning Spectral Clustering (SSC) is used to the representative samples to do the clustering. As a result, the proposed algorithm can quickly and precisely realize the clustering for texture image segmentation.

AB - Clustering is a popular and effective method for texture image segmentation. However, most cluster methods often suffer the following problems: need a huge space and a lot of computation when the input data is large. To save the space and computation, we construct a novel algorithm for image segmentation. It consists of two phases: Sampling and clustering. First, we put some detectors into the data space uniformly using orthogonal design method. These detectors can move and merge according to the law of universal gravitation. When the detectors are in a stable status (i.e., do not move), these detectors are used as the representative samples to the next step. Second, to further improve the efficiency and avoid dependence on parameters, the Self-tuning Spectral Clustering (SSC) is used to the representative samples to do the clustering. As a result, the proposed algorithm can quickly and precisely realize the clustering for texture image segmentation.

KW - Feature extraction

KW - Gray level histogram

KW - Sampling

KW - Spectral clustering

KW - Texture image segmentation

UR - http://www.scopus.com/inward/record.url?scp=84951771890&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-21380-4_113

DO - 10.1007/978-3-319-21380-4_113

M3 - Conference proceeding

AN - SCOPUS:84951771890

SN - 9783319213798

T3 - Communications in Computer and Information Science

SP - 671

EP - 676

BT - HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings

A2 - Stephanidis, Constantine

PB - Springer Verlag

T2 - 17th International Conference on Human Computer Interaction, HCI 2015

Y2 - 2 August 2015 through 7 August 2015

ER -

Texture image segmentation using spectral clustering

Abstract

Publication series

Conference

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this