Nonparametric Multiscale Blind Estimation of Intensity-Frequency-Dependent Noise

Miguel Colom; Marc Lebrun; Antoni Buades; Jean-Michel Morel

doi:10.1109/TIP.2015.2438537

Nonparametric Multiscale Blind Estimation of Intensity-Frequency-Dependent Noise

Miguel Colom
, Marc Lebrun
, Antoni Buades
, Jean-Michel Morel

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

22 Citations (Scopus)

Abstract

The camera calibration parameters and the image processing chain which generated a given image are generally not available to the receiver. This happens for example with scanned photographs and for most JPEG images. These images have undergone various nonlinear contrast changes and also linear and nonlinear filters. To deal with remnant noise in such images, we introduce a general nonparametric intensity and frequency-dependent noise model. We demonstrate by simulated and experiments with real images that this model, which requires the estimation of more than 1000 parameters, performs an efficient noise estimation. The proposed noise model is a patch model. Its estimation can therefore be used as a preliminary step to any patch-based denoising method. Our noise estimation method introduces several new tools for performing this complex estimation. One of them is a new sparse patch distance function permitting to find noisy patches with similar underlying geometry. A validation of the noise model and of its estimation method is obtained by comparing its results to ground-truth noise curves for both raw and JPEG-encoded images, and by visual inspection of the denoising results of real images. A fair comparison with the state of the art is also performed. © 1992-2012 IEEE.

Original language	English
Article number	7113861
Pages (from-to)	3162-3175
Journal	IEEE Transactions on Image Processing
Volume	24
Issue number	10
DOIs	https://doi.org/10.1109/TIP.2015.2438537
Publication status	Published - 1 Oct 2015
Externally published	Yes

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 was supported in part by the Agence Nationale de la Recherche–Direction Générale de l’Armement (DGA) through the Stéréo Project under Grant ANR-12-ASTR-0035, in part by the Fondation Mathématique Jacques Hadamard, in part by the European Research Council through the Advanced Grant Twelve Labours, in part by DGA, in part by the Centre National d’Etudes Spatiales through MISS Project, in part by the Office of Naval Research, Arlington, VA, USA, under Grant N00014-14-1-0023, in part by the Spanish Ministerio de Economía y Competitividad under Grant TIN2011-27539, in part by Fonds Unique Interministériel through the Project Plein Phare, and in part by the Institut Universitaire de France.

Research Keywords

Blind denoising
Blind noise estimation
Frequency-dependent noise
Multiscale estimation
Nonparametric noise model
Signal-dependent noise

Access Output

10.1109/TIP.2015.2438537

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{e8d69c3b307a4fa69e4bc2ba42b4778b,

title = "Nonparametric Multiscale Blind Estimation of Intensity-Frequency-Dependent Noise",

abstract = "The camera calibration parameters and the image processing chain which generated a given image are generally not available to the receiver. This happens for example with scanned photographs and for most JPEG images. These images have undergone various nonlinear contrast changes and also linear and nonlinear filters. To deal with remnant noise in such images, we introduce a general nonparametric intensity and frequency-dependent noise model. We demonstrate by simulated and experiments with real images that this model, which requires the estimation of more than 1000 parameters, performs an efficient noise estimation. The proposed noise model is a patch model. Its estimation can therefore be used as a preliminary step to any patch-based denoising method. Our noise estimation method introduces several new tools for performing this complex estimation. One of them is a new sparse patch distance function permitting to find noisy patches with similar underlying geometry. A validation of the noise model and of its estimation method is obtained by comparing its results to ground-truth noise curves for both raw and JPEG-encoded images, and by visual inspection of the denoising results of real images. A fair comparison with the state of the art is also performed. {\textcopyright} 1992-2012 IEEE.",

keywords = "Blind denoising, Blind noise estimation, Frequency-dependent noise, Multiscale estimation, Nonparametric noise model, Signal-dependent noise",

author = "Miguel Colom and Marc Lebrun and Antoni Buades and Jean-Michel Morel",

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 = "2015",

month = oct,

day = "1",

doi = "10.1109/TIP.2015.2438537",

language = "English",

volume = "24",

pages = "3162--3175",

journal = "IEEE Transactions on Image Processing",

issn = "1057-7149",

publisher = "IEEE",

number = "10",

}

TY - JOUR

T1 - Nonparametric Multiscale Blind Estimation of Intensity-Frequency-Dependent Noise

AU - Colom, Miguel

AU - Lebrun, Marc

AU - Buades, Antoni

AU - Morel, Jean-Michel

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 - 2015/10/1

Y1 - 2015/10/1

N2 - The camera calibration parameters and the image processing chain which generated a given image are generally not available to the receiver. This happens for example with scanned photographs and for most JPEG images. These images have undergone various nonlinear contrast changes and also linear and nonlinear filters. To deal with remnant noise in such images, we introduce a general nonparametric intensity and frequency-dependent noise model. We demonstrate by simulated and experiments with real images that this model, which requires the estimation of more than 1000 parameters, performs an efficient noise estimation. The proposed noise model is a patch model. Its estimation can therefore be used as a preliminary step to any patch-based denoising method. Our noise estimation method introduces several new tools for performing this complex estimation. One of them is a new sparse patch distance function permitting to find noisy patches with similar underlying geometry. A validation of the noise model and of its estimation method is obtained by comparing its results to ground-truth noise curves for both raw and JPEG-encoded images, and by visual inspection of the denoising results of real images. A fair comparison with the state of the art is also performed. © 1992-2012 IEEE.

AB - The camera calibration parameters and the image processing chain which generated a given image are generally not available to the receiver. This happens for example with scanned photographs and for most JPEG images. These images have undergone various nonlinear contrast changes and also linear and nonlinear filters. To deal with remnant noise in such images, we introduce a general nonparametric intensity and frequency-dependent noise model. We demonstrate by simulated and experiments with real images that this model, which requires the estimation of more than 1000 parameters, performs an efficient noise estimation. The proposed noise model is a patch model. Its estimation can therefore be used as a preliminary step to any patch-based denoising method. Our noise estimation method introduces several new tools for performing this complex estimation. One of them is a new sparse patch distance function permitting to find noisy patches with similar underlying geometry. A validation of the noise model and of its estimation method is obtained by comparing its results to ground-truth noise curves for both raw and JPEG-encoded images, and by visual inspection of the denoising results of real images. A fair comparison with the state of the art is also performed. © 1992-2012 IEEE.

KW - Blind denoising

KW - Blind noise estimation

KW - Frequency-dependent noise

KW - Multiscale estimation

KW - Nonparametric noise model

KW - Signal-dependent noise

UR - https://www.scopus.com/pages/publications/84933074089

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84933074089&origin=recordpage

U2 - 10.1109/TIP.2015.2438537

DO - 10.1109/TIP.2015.2438537

M3 - RGC 21 - Publication in refereed journal

SN - 1057-7149

VL - 24

SP - 3162

EP - 3175

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

IS - 10

M1 - 7113861

ER -

Nonparametric Multiscale Blind Estimation of Intensity-Frequency-Dependent Noise

Abstract

Bibliographical note

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this