Speckle noise reduction for digital holographic images using multi-scale convolutional neural networks

Wonseok Jeon; Wooyoung Jeong; Kyungchan Son; Hyunseok Yang

doi:10.1364/OL.43.004240

Speckle noise reduction for digital holographic images using multi-scale convolutional neural networks

Wonseok Jeon, Wooyoung Jeong, Kyungchan Son, Hyunseok Yang

Department of Mechanical Engineering

Research output: Contribution to journal › Article › peer-review

23 Citations (Scopus)

Abstract

In this Letter, we propose a fast speckle noise reduction method with only a single reconstructed image based on convolutional neural networks. The proposed network has multi-sized kernels that can capture the speckle noise component effectively from digital holographic images. For robust noise reduction performance, the network is trained with a large noisy image dataset that has object-dependent noise and a wide range of noise levels. The experimental results show the fast, robust, and outstanding speckle noise reduction performance of the proposed approach.

Original language	English
Pages (from-to)	4240-4243
Number of pages	4
Journal	Optics Letters
Volume	43
Issue number	17
DOIs	https://doi.org/10.1364/OL.43.004240
Publication status	Published - 2018 Sep 1

Bibliographical note

Funding Information:
Ministry of Education (MOE) (NRF-2016R1D1A1B03930730). This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education.

Funding Information:
Acknowledgment. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education.

All Science Journal Classification (ASJC) codes

Atomic and Molecular Physics, and Optics

Access to Document

10.1364/OL.43.004240

Fingerprint Dive into the research topics of 'Speckle noise reduction for digital holographic images using multi-scale convolutional neural networks'. Together they form a unique fingerprint.

Cite this

@article{c35bc608c6a04df897bcafc18c5e2087,

title = "Speckle noise reduction for digital holographic images using multi-scale convolutional neural networks",

abstract = "In this Letter, we propose a fast speckle noise reduction method with only a single reconstructed image based on convolutional neural networks. The proposed network has multi-sized kernels that can capture the speckle noise component effectively from digital holographic images. For robust noise reduction performance, the network is trained with a large noisy image dataset that has object-dependent noise and a wide range of noise levels. The experimental results show the fast, robust, and outstanding speckle noise reduction performance of the proposed approach.",

author = "Wonseok Jeon and Wooyoung Jeong and Kyungchan Son and Hyunseok Yang",

note = "Funding Information: Ministry of Education (MOE) (NRF-2016R1D1A1B03930730). This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education. Funding Information: Acknowledgment. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education.",

year = "2018",

month = sep,

day = "1",

doi = "10.1364/OL.43.004240",

language = "English",

volume = "43",

pages = "4240--4243",

journal = "Optics Letters",

issn = "0146-9592",

publisher = "The Optical Society",

number = "17",

}

TY - JOUR

T1 - Speckle noise reduction for digital holographic images using multi-scale convolutional neural networks

AU - Jeon, Wonseok

AU - Jeong, Wooyoung

AU - Son, Kyungchan

AU - Yang, Hyunseok

N1 - Funding Information: Ministry of Education (MOE) (NRF-2016R1D1A1B03930730). This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education. Funding Information: Acknowledgment. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education.

PY - 2018/9/1

Y1 - 2018/9/1

N2 - In this Letter, we propose a fast speckle noise reduction method with only a single reconstructed image based on convolutional neural networks. The proposed network has multi-sized kernels that can capture the speckle noise component effectively from digital holographic images. For robust noise reduction performance, the network is trained with a large noisy image dataset that has object-dependent noise and a wide range of noise levels. The experimental results show the fast, robust, and outstanding speckle noise reduction performance of the proposed approach.

AB - In this Letter, we propose a fast speckle noise reduction method with only a single reconstructed image based on convolutional neural networks. The proposed network has multi-sized kernels that can capture the speckle noise component effectively from digital holographic images. For robust noise reduction performance, the network is trained with a large noisy image dataset that has object-dependent noise and a wide range of noise levels. The experimental results show the fast, robust, and outstanding speckle noise reduction performance of the proposed approach.

UR - http://www.scopus.com/inward/record.url?scp=85052578601&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052578601&partnerID=8YFLogxK

U2 - 10.1364/OL.43.004240

DO - 10.1364/OL.43.004240

M3 - Article

C2 - 30160761

AN - SCOPUS:85052578601

VL - 43

SP - 4240

EP - 4243

JO - Optics Letters

JF - Optics Letters

SN - 0146-9592

IS - 17

ER -