A novel matrix norm based Gaussian kernel for feature extraction of images

Jun Bao Li; Shu Chuan Chu; Jeng Shyang Pan; Jiun Huei Ho

doi:10.1109/IIH-MSP.2006.265004

A novel matrix norm based Gaussian kernel for feature extraction of images

Jun Bao Li^*
, Shu Chuan Chu
, Jeng Shyang Pan
, Jiun Huei Ho

^*Corresponding author for this work

Faculty of Computing

Cheng Shiu University Taiwan
National Kaohsiung University of Science and Technology
Cheng Shiu University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

Gaussian kernel is widely used in Support Vector Machines and many other kernel methods, and it is most often deemed to provide a local measure of similarity between vectors, which causes large storage requirements and large computational effort for transforming images to vectors owing to its viewing images as vectors. A novel matrix norm based Gaussian kernel (M-Gaussian kernel) which views images as matrices is proposed to solve the problem. Experiments conducted on ORL face database show the effectiveness of the proposed M-Gaussian kernel.

Original language	English
Title of host publication	Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006
Pages	305-308
Number of pages	4
DOIs	https://doi.org/10.1109/IIH-MSP.2006.265004
State	Published - 2006
Event	2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006 - Pasadena, CA, United States Duration: 18 Dec 2006 → 20 Dec 2006

Publication series

Name	Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006

Conference

Conference	2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006
Country/Territory	United States
City	Pasadena, CA
Period	18/12/06 → 20/12/06

Access to Document

10.1109/IIH-MSP.2006.265004

Cite this

Li, J. B., Chu, S. C., Pan, J. S., & Ho, J. H. (2006). A novel matrix norm based Gaussian kernel for feature extraction of images. In Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006 (pp. 305-308). Article 4041724 (Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006). https://doi.org/10.1109/IIH-MSP.2006.265004

Li, Jun Bao ; Chu, Shu Chuan ; Pan, Jeng Shyang et al. / A novel matrix norm based Gaussian kernel for feature extraction of images. Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006. 2006. pp. 305-308 (Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006).

@inproceedings{ec1225e0528548069b6a3bd0ded1c3a7,

title = "A novel matrix norm based Gaussian kernel for feature extraction of images",

abstract = "Gaussian kernel is widely used in Support Vector Machines and many other kernel methods, and it is most often deemed to provide a local measure of similarity between vectors, which causes large storage requirements and large computational effort for transforming images to vectors owing to its viewing images as vectors. A novel matrix norm based Gaussian kernel (M-Gaussian kernel) which views images as matrices is proposed to solve the problem. Experiments conducted on ORL face database show the effectiveness of the proposed M-Gaussian kernel.",

author = "Li, \{Jun Bao\} and Chu, \{Shu Chuan\} and Pan, \{Jeng Shyang\} and Ho, \{Jiun Huei\}",

year = "2006",

doi = "10.1109/IIH-MSP.2006.265004",

language = "英语",

isbn = "0769527450",

series = "Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006",

pages = "305--308",

booktitle = "Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006",

note = "2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006 ; Conference date: 18-12-2006 Through 20-12-2006",

}

Li, JB, Chu, SC, Pan, JS & Ho, JH 2006, A novel matrix norm based Gaussian kernel for feature extraction of images. in Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006., 4041724, Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006, pp. 305-308, 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006, Pasadena, CA, United States, 18/12/06. https://doi.org/10.1109/IIH-MSP.2006.265004

A novel matrix norm based Gaussian kernel for feature extraction of images. / Li, Jun Bao; Chu, Shu Chuan; Pan, Jeng Shyang et al.
Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006. 2006. p. 305-308 4041724 (Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A novel matrix norm based Gaussian kernel for feature extraction of images

AU - Li, Jun Bao

AU - Chu, Shu Chuan

AU - Pan, Jeng Shyang

AU - Ho, Jiun Huei

PY - 2006

Y1 - 2006

N2 - Gaussian kernel is widely used in Support Vector Machines and many other kernel methods, and it is most often deemed to provide a local measure of similarity between vectors, which causes large storage requirements and large computational effort for transforming images to vectors owing to its viewing images as vectors. A novel matrix norm based Gaussian kernel (M-Gaussian kernel) which views images as matrices is proposed to solve the problem. Experiments conducted on ORL face database show the effectiveness of the proposed M-Gaussian kernel.

AB - Gaussian kernel is widely used in Support Vector Machines and many other kernel methods, and it is most often deemed to provide a local measure of similarity between vectors, which causes large storage requirements and large computational effort for transforming images to vectors owing to its viewing images as vectors. A novel matrix norm based Gaussian kernel (M-Gaussian kernel) which views images as matrices is proposed to solve the problem. Experiments conducted on ORL face database show the effectiveness of the proposed M-Gaussian kernel.

UR - https://www.scopus.com/pages/publications/40349100579

U2 - 10.1109/IIH-MSP.2006.265004

DO - 10.1109/IIH-MSP.2006.265004

M3 - 会议稿件

AN - SCOPUS:40349100579

SN - 0769527450

SN - 9780769527451

T3 - Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006

SP - 305

EP - 308

BT - Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006

T2 - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006

Y2 - 18 December 2006 through 20 December 2006

ER -

Li JB, Chu SC, Pan JS, Ho JH. A novel matrix norm based Gaussian kernel for feature extraction of images. In Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006. 2006. p. 305-308. 4041724. (Proceedings - 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2006). doi: 10.1109/IIH-MSP.2006.265004

A novel matrix norm based Gaussian kernel for feature extraction of images

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this