Asymmetric, non-unimodal kernel regression for image processing

Damith J. Mudugamuwa; Wenjing Jia; Xiangjian He

doi:10.1109/DICTA.2010.34

Asymmetric, non-unimodal kernel regression for image processing

Damith J. Mudugamuwa, Wenjing Jia, Xiangjian He

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

Abstract

Kernel regression has been previously proposed as a robust estimator for a wide range of image processing tasks, including image denoising, interpolation and super-resolution. In this article we propose a kernel formulation that relaxes the usual symmetric and unimodal properties to effectively exploit the smoothness characteristics of natural images. The proposed method extends the kernel support along similar image characteristics to further increase the robustness of the estimates. Application of the proposed method to image denoising yields significant improvement over the previously reported regression methods and produces results comparable to the state-of-the-art denoising techniques.

Original language	English
Title of host publication	Proceedings - 2010 Digital Image Computing
Subtitle of host publication	Techniques and Applications, DICTA 2010
Pages	141-145
Number of pages	5
DOIs	https://doi.org/10.1109/DICTA.2010.34
Publication status	Published - 2010
Externally published	Yes
Event	International Conference on Digital Image Computing: Techniques and Applications, DICTA 2010 - Sydney, NSW, Australia Duration: 1 Dec 2010 → 3 Dec 2010

Publication series

Name	Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010

Conference

Conference	International Conference on Digital Image Computing: Techniques and Applications, DICTA 2010
Country/Territory	Australia
City	Sydney, NSW
Period	1/12/10 → 3/12/10

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/DICTA.2010.34

Cite this

@inproceedings{7b480089ccf44c11a5fac8def8b679cd,

title = "Asymmetric, non-unimodal kernel regression for image processing",

abstract = "Kernel regression has been previously proposed as a robust estimator for a wide range of image processing tasks, including image denoising, interpolation and super-resolution. In this article we propose a kernel formulation that relaxes the usual symmetric and unimodal properties to effectively exploit the smoothness characteristics of natural images. The proposed method extends the kernel support along similar image characteristics to further increase the robustness of the estimates. Application of the proposed method to image denoising yields significant improvement over the previously reported regression methods and produces results comparable to the state-of-the-art denoising techniques.",

author = "Mudugamuwa, {Damith J.} and Wenjing Jia and Xiangjian He",

year = "2010",

doi = "10.1109/DICTA.2010.34",

language = "English",

isbn = "9780769542713",

series = "Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010",

pages = "141--145",

booktitle = "Proceedings - 2010 Digital Image Computing",

note = "International Conference on Digital Image Computing: Techniques and Applications, DICTA 2010 ; Conference date: 01-12-2010 Through 03-12-2010",

}

Mudugamuwa, DJ, Jia, W & He, X 2010, Asymmetric, non-unimodal kernel regression for image processing. in Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010., 5692555, Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010, pp. 141-145, International Conference on Digital Image Computing: Techniques and Applications, DICTA 2010, Sydney, NSW, Australia, 1/12/10. https://doi.org/10.1109/DICTA.2010.34

Asymmetric, non-unimodal kernel regression for image processing. / Mudugamuwa, Damith J.; Jia, Wenjing; He, Xiangjian.
Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010. 2010. p. 141-145 5692555 (Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010).

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

TY - GEN

T1 - Asymmetric, non-unimodal kernel regression for image processing

AU - Mudugamuwa, Damith J.

AU - Jia, Wenjing

AU - He, Xiangjian

PY - 2010

Y1 - 2010

N2 - Kernel regression has been previously proposed as a robust estimator for a wide range of image processing tasks, including image denoising, interpolation and super-resolution. In this article we propose a kernel formulation that relaxes the usual symmetric and unimodal properties to effectively exploit the smoothness characteristics of natural images. The proposed method extends the kernel support along similar image characteristics to further increase the robustness of the estimates. Application of the proposed method to image denoising yields significant improvement over the previously reported regression methods and produces results comparable to the state-of-the-art denoising techniques.

AB - Kernel regression has been previously proposed as a robust estimator for a wide range of image processing tasks, including image denoising, interpolation and super-resolution. In this article we propose a kernel formulation that relaxes the usual symmetric and unimodal properties to effectively exploit the smoothness characteristics of natural images. The proposed method extends the kernel support along similar image characteristics to further increase the robustness of the estimates. Application of the proposed method to image denoising yields significant improvement over the previously reported regression methods and produces results comparable to the state-of-the-art denoising techniques.

UR - http://www.scopus.com/inward/record.url?scp=79951642846&partnerID=8YFLogxK

U2 - 10.1109/DICTA.2010.34

DO - 10.1109/DICTA.2010.34

M3 - Conference contribution

AN - SCOPUS:79951642846

SN - 9780769542713

T3 - Proceedings - 2010 Digital Image Computing: Techniques and Applications, DICTA 2010

SP - 141

EP - 145

BT - Proceedings - 2010 Digital Image Computing

T2 - International Conference on Digital Image Computing: Techniques and Applications, DICTA 2010

Y2 - 1 December 2010 through 3 December 2010

ER -

Asymmetric, non-unimodal kernel regression for image processing

Abstract

Publication series

Conference

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this