Can a single image denoising neural network handle all levels of gaussian noise

Yi-Qing Wang; Jean-Michel Morel

doi:10.1109/LSP.2014.2314613

Author(s)

Yi-Qing Wang
Jean-Michel Morel

Detail(s)

Original language	English
Article number	6781616
Pages (from-to)	1150-1153
Journal / Publication	IEEE Signal Processing Letters
Volume	21
Issue number	9
Publication status	Published - Sept 2014
Externally published	Yes

Link(s)

DOI	DOI https://doi.org/10.1109/LSP.2014.2314613 Final Published version
	Check@CityULib
Link to Scopus	https://www.scopus.com/record/display.uri?eid=2-s2.0-84901984291&origin=recordpage
Permanent Link	https://scholars.cityu.edu.hk/en/publications/publication(c8cacda2-2225-402f-b4ab-751b5f8b9b98).html

Abstract

A recently introduced set of deep neural networks designed for the image denoising task achieves state-of-the-art performance. However, they are specialized networks in that each of them can handle just one noise level fixed in their respective training process. In this letter, by investigating the distribution invariance of the natural image patches with respect to linear transforms, we show how to make a single existing deep neural network work well across all levels of Gaussian noise, thereby allowing to significantly reduce the training time for a general-purpose neural network powered denoising algorithm. © 2014 IEEE.

Research Area(s)

Deep neural network, distribution invariance, image denoising, natural patch space

Bibliographic 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].

Citation Format(s)

[ RIS ] [ BibTeX ]

Can a single image denoising neural network handle all levels of gaussian noise. / Wang, Yi-Qing; Morel, Jean-Michel.
In: IEEE Signal Processing Letters, Vol. 21, No. 9, 6781616, 09.2014, p. 1150-1153.

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

Wang, Y-Q & Morel, J-M 2014, 'Can a single image denoising neural network handle all levels of gaussian noise', IEEE Signal Processing Letters, vol. 21, no. 9, 6781616, pp. 1150-1153. https://doi.org/10.1109/LSP.2014.2314613

Wang, Y.-Q., & Morel, J.-M. (2014). Can a single image denoising neural network handle all levels of gaussian noise. IEEE Signal Processing Letters, 21(9), 1150-1153. Article 6781616. https://doi.org/10.1109/LSP.2014.2314613

Wang YQ, Morel JM. Can a single image denoising neural network handle all levels of gaussian noise. IEEE Signal Processing Letters. 2014 Sept;21(9):1150-1153. 6781616. doi: 10.1109/LSP.2014.2314613

Wang, Yi-Qing ; Morel, Jean-Michel. / Can a single image denoising neural network handle all levels of gaussian noise. In: IEEE Signal Processing Letters. 2014 ; Vol. 21, No. 9. pp. 1150-1153.