An L 1 -based variational model for Retinex theory and its application to medical images

Wenye Ma; Jean-Michel Morel; Stanley Osher; Aichi Chien

doi:10.1109/CVPR.2011.5995422

An L ₁ -based variational model for Retinex theory and its application to medical images

Wenye Ma
, Jean-Michel Morel
, Stanley Osher
, Aichi Chien

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

118 Citations (Scopus)

Abstract

Human visual system (HVS) can perceive constant color under varying illumination conditions while digital images record information of both reflectance (physical color) of objects and illumination. Retinex theory, formulated by Edwin H. Land, aimed to simulate and explain this feature of HVS. However, to recover the reflectance from a given image is in general an ill-posed problem. In this paper, we establish an L ₁ -based variational model for Retinex theory that can be solved by a fast computational approach based on Bregman iteration. Compared with previous works, our L ₁ -Retinex method is more accurate for recovering the reflectance, which is illustrated by examples and statistics. In medical images such as magnetic resonance imaging (MRI), intensity inhomogeneity is often encountered due to bias fields. This is a similar formulation to Retinex theory while the MRI has some specific properties. We then modify the L ₁ -Retinex method and develop a new algorithm for MRI data. We demonstrate the performance of our method by comparison with previous work on simulated and real data. © 2011 IEEE.

Original language	English
Title of host publication	2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011
Publisher	IEEE Computer Society
Pages	153-160
ISBN (Print)	9781457703942
DOIs	https://doi.org/10.1109/CVPR.2011.5995422
Publication status	Published - 2011
Externally published	Yes

Publication series

Name	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN (Print)	1063-6919

Bibliographical note

Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

Funding

This work is partially supported by NSF DMS 0835863, DMS 0914561, DMS 0914856, CCF 0830554, ONR N000 14–08-1119, N000 14–10–1 0221, and the Department of Defense.

Access Output

10.1109/CVPR.2011.5995422

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Ma, W., Morel, J.-M., Osher, S., & Chien, A. (2011). An L ₁ -based variational model for Retinex theory and its application to medical images. In 2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011 (pp. 153-160). Article 5995422 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). IEEE Computer Society. https://doi.org/10.1109/CVPR.2011.5995422

@inproceedings{65de6efbcef9451fae1e8fa943a4889f,

title = "An L 1 -based variational model for Retinex theory and its application to medical images",

abstract = "Human visual system (HVS) can perceive constant color under varying illumination conditions while digital images record information of both reflectance (physical color) of objects and illumination. Retinex theory, formulated by Edwin H. Land, aimed to simulate and explain this feature of HVS. However, to recover the reflectance from a given image is in general an ill-posed problem. In this paper, we establish an L 1 -based variational model for Retinex theory that can be solved by a fast computational approach based on Bregman iteration. Compared with previous works, our L 1 -Retinex method is more accurate for recovering the reflectance, which is illustrated by examples and statistics. In medical images such as magnetic resonance imaging (MRI), intensity inhomogeneity is often encountered due to bias fields. This is a similar formulation to Retinex theory while the MRI has some specific properties. We then modify the L 1 -Retinex method and develop a new algorithm for MRI data. We demonstrate the performance of our method by comparison with previous work on simulated and real data. {\textcopyright} 2011 IEEE.",

author = "Wenye Ma and Jean-Michel Morel and Stanley Osher and Aichi Chien",

note = "Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected]. ",

year = "2011",

doi = "10.1109/CVPR.2011.5995422",

language = "English",

isbn = "9781457703942",

series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

publisher = "IEEE Computer Society",

pages = "153--160",

booktitle = "2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011",

address = "United States",

}

Ma, W, Morel, J-M, Osher, S & Chien, A 2011, An L ₁ -based variational model for Retinex theory and its application to medical images. in 2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011., 5995422, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, pp. 153-160. https://doi.org/10.1109/CVPR.2011.5995422

An L ₁ -based variational model for Retinex theory and its application to medical images. / Ma, Wenye; Morel, Jean-Michel; Osher, Stanley et al.
2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011. IEEE Computer Society, 2011. p. 153-160 5995422 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - An L 1 -based variational model for Retinex theory and its application to medical images

AU - Ma, Wenye

AU - Morel, Jean-Michel

AU - Osher, Stanley

AU - Chien, Aichi

N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

PY - 2011

Y1 - 2011

N2 - Human visual system (HVS) can perceive constant color under varying illumination conditions while digital images record information of both reflectance (physical color) of objects and illumination. Retinex theory, formulated by Edwin H. Land, aimed to simulate and explain this feature of HVS. However, to recover the reflectance from a given image is in general an ill-posed problem. In this paper, we establish an L 1 -based variational model for Retinex theory that can be solved by a fast computational approach based on Bregman iteration. Compared with previous works, our L 1 -Retinex method is more accurate for recovering the reflectance, which is illustrated by examples and statistics. In medical images such as magnetic resonance imaging (MRI), intensity inhomogeneity is often encountered due to bias fields. This is a similar formulation to Retinex theory while the MRI has some specific properties. We then modify the L 1 -Retinex method and develop a new algorithm for MRI data. We demonstrate the performance of our method by comparison with previous work on simulated and real data. © 2011 IEEE.

AB - Human visual system (HVS) can perceive constant color under varying illumination conditions while digital images record information of both reflectance (physical color) of objects and illumination. Retinex theory, formulated by Edwin H. Land, aimed to simulate and explain this feature of HVS. However, to recover the reflectance from a given image is in general an ill-posed problem. In this paper, we establish an L 1 -based variational model for Retinex theory that can be solved by a fast computational approach based on Bregman iteration. Compared with previous works, our L 1 -Retinex method is more accurate for recovering the reflectance, which is illustrated by examples and statistics. In medical images such as magnetic resonance imaging (MRI), intensity inhomogeneity is often encountered due to bias fields. This is a similar formulation to Retinex theory while the MRI has some specific properties. We then modify the L 1 -Retinex method and develop a new algorithm for MRI data. We demonstrate the performance of our method by comparison with previous work on simulated and real data. © 2011 IEEE.

UR - https://www.scopus.com/pages/publications/80052907477

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-80052907477&origin=recordpage

U2 - 10.1109/CVPR.2011.5995422

DO - 10.1109/CVPR.2011.5995422

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781457703942

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 153

EP - 160

BT - 2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011

PB - IEEE Computer Society

ER -

Ma W, Morel JM, Osher S, Chien A. An L ₁ -based variational model for Retinex theory and its application to medical images. In 2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011. IEEE Computer Society. 2011. p. 153-160. 5995422. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). doi: 10.1109/CVPR.2011.5995422