By the same authors

From the same journal

A Riemannian weighted filter for edge-sensitive image smoothing

Research output: Contribution to journalArticlepeer-review

Standard

A Riemannian weighted filter for edge-sensitive image smoothing. / Zhang, Fan; Hancock, Edwin R.

In: 18th International Conference on Pattern Recognition, Vol 1, Proceedings, 2006, p. 594-598.

Research output: Contribution to journalArticlepeer-review

Harvard

Zhang, F & Hancock, ER 2006, 'A Riemannian weighted filter for edge-sensitive image smoothing', 18th International Conference on Pattern Recognition, Vol 1, Proceedings, pp. 594-598.

APA

Zhang, F., & Hancock, E. R. (2006). A Riemannian weighted filter for edge-sensitive image smoothing. 18th International Conference on Pattern Recognition, Vol 1, Proceedings, 594-598.

Vancouver

Zhang F, Hancock ER. A Riemannian weighted filter for edge-sensitive image smoothing. 18th International Conference on Pattern Recognition, Vol 1, Proceedings. 2006;594-598.

Author

Zhang, Fan ; Hancock, Edwin R. / A Riemannian weighted filter for edge-sensitive image smoothing. In: 18th International Conference on Pattern Recognition, Vol 1, Proceedings. 2006 ; pp. 594-598.

Bibtex - Download

@article{fa334daac6e1481081421f3210e6b023,
title = "A Riemannian weighted filter for edge-sensitive image smoothing",
abstract = "This paper describes a new method for image smoothing. We view the image features as residing on a differential manifold, and we work with a representation based on the exponential map for this manifold (i.e. the map from the manifold to a plane that preserves geodesic distances). On the exponential map we characterise the features using a Riemannian weighted mean. We show how both gradient descent and Newton's method can be used to find the mean. Based on this weighted mean, we develop an edge-preserving filter that combines Gaussian and median filters of gray-scale images. We demonstrate our algorithm both on direction fields from shape-from-shading and tensor-valued images.",
author = "Fan Zhang and Hancock, {Edwin R.}",
year = "2006",
language = "English",
pages = "594--598",
journal = "18th International Conference on Pattern Recognition, Vol 1, Proceedings",
issn = "1051-4651",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - A Riemannian weighted filter for edge-sensitive image smoothing

AU - Zhang, Fan

AU - Hancock, Edwin R.

PY - 2006

Y1 - 2006

N2 - This paper describes a new method for image smoothing. We view the image features as residing on a differential manifold, and we work with a representation based on the exponential map for this manifold (i.e. the map from the manifold to a plane that preserves geodesic distances). On the exponential map we characterise the features using a Riemannian weighted mean. We show how both gradient descent and Newton's method can be used to find the mean. Based on this weighted mean, we develop an edge-preserving filter that combines Gaussian and median filters of gray-scale images. We demonstrate our algorithm both on direction fields from shape-from-shading and tensor-valued images.

AB - This paper describes a new method for image smoothing. We view the image features as residing on a differential manifold, and we work with a representation based on the exponential map for this manifold (i.e. the map from the manifold to a plane that preserves geodesic distances). On the exponential map we characterise the features using a Riemannian weighted mean. We show how both gradient descent and Newton's method can be used to find the mean. Based on this weighted mean, we develop an edge-preserving filter that combines Gaussian and median filters of gray-scale images. We demonstrate our algorithm both on direction fields from shape-from-shading and tensor-valued images.

M3 - Article

SP - 594

EP - 598

JO - 18th International Conference on Pattern Recognition, Vol 1, Proceedings

JF - 18th International Conference on Pattern Recognition, Vol 1, Proceedings

SN - 1051-4651

ER -