Detection of Gaussian noise and its level using deep Convolutional Neural Network

Joon Huang Chuah; Hui Ying Khaw; Foo Chong Soon; Chee Onn Chow

doi:10.1109/TENCON.2017.8228272

Detection of Gaussian noise and its level using deep Convolutional Neural Network

Joon Huang Chuah
, Hui Ying Khaw
, Foo Chong Soon
, Chee Onn Chow

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research

39 Citations (Scopus)

Abstract

This study presents a Convolutional Neural Network (CNN) model to effectively recognize the presence of Gaussian noise and its level in images. The existing denoising approaches are mostly based on an assumption that the images to be processed are corrupted with noises. This work, on the other hand, aims to intelligently evaluate if an image is corrupted, and to which level it is degraded, before applying denoising algorithms. We used 12000 and 3000 standard test images for training and testing purposes, respectively. Different noise levels are introduced to these images. The overall accuracy of 74.7% in classifying 10 classes of noise levels are obtained. Our experiments and results have proven that this model is capable of performing Gaussian noise detection and its noise level classification.

Original language	English
Title of host publication	Proceedings of the 2017 IEEE Region 10 Conference (TENCON)
Editors	Mohammad Faizal Ahmad Fauzi
Place of Publication	Piscataway NJ USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	2447-2450
Number of pages	4
ISBN (Electronic)	9781509011339, 9781509011346
ISBN (Print)	9781509011353
DOIs	https://doi.org/10.1109/TENCON.2017.8228272
Publication status	Published - 2017
Externally published	Yes
Event	IEEE Tencon (IEEE Region 10 Conference) 2017 - Penang, Malaysia Duration: 5 Nov 2017 → 8 Nov 2017 https://ieeexplore.ieee.org/xpl/conhome/8169968/proceeding (Proceedings) https://ieeemy.org/tencon/ (Website)

Publication series

Name	IEEE Region 10 Annual International Conference, Proceedings/TENCON
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Volume	2017-December
ISSN (Print)	2159-3442
ISSN (Electronic)	2159-3450

Conference

Conference	IEEE Tencon (IEEE Region 10 Conference) 2017
Abbreviated title	TENCON 2017
Country/Territory	Malaysia
City	Penang
Period	5/11/17 → 8/11/17
Internet address	https://ieeexplore.ieee.org/xpl/conhome/8169968/proceeding (Proceedings) https://ieeemy.org/tencon/ (Website)

Keywords

convolutional neural networks
Gaussian noise
image noise
noise detection
training

Access to Document

10.1109/TENCON.2017.8228272

Cite this

Chuah, J. H., Khaw, H. Y., Soon, F. C., & Chow, C. O. (2017). Detection of Gaussian noise and its level using deep Convolutional Neural Network. In M. Faizal Ahmad Fauzi (Ed.), Proceedings of the 2017 IEEE Region 10 Conference (TENCON) (pp. 2447-2450). (IEEE Region 10 Annual International Conference, Proceedings/TENCON; Vol. 2017-December). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/TENCON.2017.8228272

Chuah, Joon Huang ; Khaw, Hui Ying ; Soon, Foo Chong et al. / Detection of Gaussian noise and its level using deep Convolutional Neural Network. Proceedings of the 2017 IEEE Region 10 Conference (TENCON). editor / Mohammad Faizal Ahmad Fauzi. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 2447-2450 (IEEE Region 10 Annual International Conference, Proceedings/TENCON).

@inproceedings{a7093cb5ab4a49948546ef962db07fb7,

title = "Detection of Gaussian noise and its level using deep Convolutional Neural Network",

abstract = "This study presents a Convolutional Neural Network (CNN) model to effectively recognize the presence of Gaussian noise and its level in images. The existing denoising approaches are mostly based on an assumption that the images to be processed are corrupted with noises. This work, on the other hand, aims to intelligently evaluate if an image is corrupted, and to which level it is degraded, before applying denoising algorithms. We used 12000 and 3000 standard test images for training and testing purposes, respectively. Different noise levels are introduced to these images. The overall accuracy of 74.7\% in classifying 10 classes of noise levels are obtained. Our experiments and results have proven that this model is capable of performing Gaussian noise detection and its noise level classification.",

keywords = "convolutional neural networks, Gaussian noise, image noise, noise detection, training",

author = "Chuah, \{Joon Huang\} and Khaw, \{Hui Ying\} and Soon, \{Foo Chong\} and Chow, \{Chee Onn\}",

note = "Funding Information: This work was supported by the Fundamental Research Grant Scheme (FRGS) grant from Malaysian Ministry of Higher Education [FRGS/1/2016/ICT03/UM/02/1]. Publisher Copyright: {\textcopyright} 2017 IEEE. Copyright: Copyright 2018 Elsevier B.V., All rights reserved.; IEEE Tencon (IEEE Region 10 Conference) 2017, TENCON 2017 ; Conference date: 05-11-2017 Through 08-11-2017",

year = "2017",

doi = "10.1109/TENCON.2017.8228272",

language = "English",

isbn = "9781509011353",

series = "IEEE Region 10 Annual International Conference, Proceedings/TENCON",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

pages = "2447--2450",

editor = "\{Faizal Ahmad Fauzi\}, Mohammad",

booktitle = "Proceedings of the 2017 IEEE Region 10 Conference (TENCON)",

address = "United States of America",

url = "https://ieeexplore.ieee.org/xpl/conhome/8169968/proceeding, https://ieeemy.org/tencon/",

}

Chuah, JH, Khaw, HY, Soon, FC & Chow, CO 2017, Detection of Gaussian noise and its level using deep Convolutional Neural Network. in M Faizal Ahmad Fauzi (ed.), Proceedings of the 2017 IEEE Region 10 Conference (TENCON). IEEE Region 10 Annual International Conference, Proceedings/TENCON, vol. 2017-December, IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 2447-2450, IEEE Tencon (IEEE Region 10 Conference) 2017, Penang, Malaysia, 5/11/17. https://doi.org/10.1109/TENCON.2017.8228272

Detection of Gaussian noise and its level using deep Convolutional Neural Network. / Chuah, Joon Huang; Khaw, Hui Ying; Soon, Foo Chong et al.
Proceedings of the 2017 IEEE Region 10 Conference (TENCON). ed. / Mohammad Faizal Ahmad Fauzi. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 2447-2450 (IEEE Region 10 Annual International Conference, Proceedings/TENCON; Vol. 2017-December).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research

TY - GEN

T1 - Detection of Gaussian noise and its level using deep Convolutional Neural Network

AU - Chuah, Joon Huang

AU - Khaw, Hui Ying

AU - Soon, Foo Chong

AU - Chow, Chee Onn

N1 - Funding Information: This work was supported by the Fundamental Research Grant Scheme (FRGS) grant from Malaysian Ministry of Higher Education [FRGS/1/2016/ICT03/UM/02/1]. Publisher Copyright: © 2017 IEEE. Copyright: Copyright 2018 Elsevier B.V., All rights reserved.

PY - 2017

Y1 - 2017

N2 - This study presents a Convolutional Neural Network (CNN) model to effectively recognize the presence of Gaussian noise and its level in images. The existing denoising approaches are mostly based on an assumption that the images to be processed are corrupted with noises. This work, on the other hand, aims to intelligently evaluate if an image is corrupted, and to which level it is degraded, before applying denoising algorithms. We used 12000 and 3000 standard test images for training and testing purposes, respectively. Different noise levels are introduced to these images. The overall accuracy of 74.7% in classifying 10 classes of noise levels are obtained. Our experiments and results have proven that this model is capable of performing Gaussian noise detection and its noise level classification.

AB - This study presents a Convolutional Neural Network (CNN) model to effectively recognize the presence of Gaussian noise and its level in images. The existing denoising approaches are mostly based on an assumption that the images to be processed are corrupted with noises. This work, on the other hand, aims to intelligently evaluate if an image is corrupted, and to which level it is degraded, before applying denoising algorithms. We used 12000 and 3000 standard test images for training and testing purposes, respectively. Different noise levels are introduced to these images. The overall accuracy of 74.7% in classifying 10 classes of noise levels are obtained. Our experiments and results have proven that this model is capable of performing Gaussian noise detection and its noise level classification.

KW - convolutional neural networks

KW - Gaussian noise

KW - image noise

KW - noise detection

KW - training

UR - https://www.scopus.com/pages/publications/85044223847

U2 - 10.1109/TENCON.2017.8228272

DO - 10.1109/TENCON.2017.8228272

M3 - Conference Paper

AN - SCOPUS:85044223847

SN - 9781509011353

T3 - IEEE Region 10 Annual International Conference, Proceedings/TENCON

SP - 2447

EP - 2450

BT - Proceedings of the 2017 IEEE Region 10 Conference (TENCON)

A2 - Faizal Ahmad Fauzi, Mohammad

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

T2 - IEEE Tencon (IEEE Region 10 Conference) 2017

Y2 - 5 November 2017 through 8 November 2017

ER -

Chuah JH, Khaw HY, Soon FC, Chow CO. Detection of Gaussian noise and its level using deep Convolutional Neural Network. In Faizal Ahmad Fauzi M, editor, Proceedings of the 2017 IEEE Region 10 Conference (TENCON). Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 2447-2450. (IEEE Region 10 Annual International Conference, Proceedings/TENCON). doi: 10.1109/TENCON.2017.8228272

Detection of Gaussian noise and its level using deep Convolutional Neural Network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this