Presentation 2018-12-07
A Switching Noise Removal Filter Based on Convolutional Neural Networks and Its Application to Random-valued Impulse Noise
Yukiya Fukuda, Ryosuke Kubota,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to remove random-valued impulse noise (RVIN) on a color image, we propose a novel switching denoising filter based on convolutional neural networks (CNNs). The proposed method uses two kinds of CNNs to detect and remove RVIN. In the detection process, the CNNs of the proposed method outputs a probability that a target pixel is corrupted by RVIN. In the removal process, the value of pixel detected as the RVIN based on the probability is restored to the value predicted by CNNs. The validity and effectiveness of the proposed method are verified by applying it to the noise removal of color images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Convolutional Neural Networks / Denoising / Switching Denoising Filter / Random-valued Impulse Noise
Paper # SIS2018-29
Date of Issue 2018-11-29 (SIS)

Conference Information
Committee SIS
Conference Date 2018/12/6(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Hagi Civic Center
Topics (in Japanese) (See Japanese page)
Topics (in English) Smart Personal Systems, etc.
Chair Takayuki Nakachi(NTT)
Vice Chair Noriaki Suetake(Yamaguchi Univ.) / Tomoaki Kimura(Kanagawa Inst. of Tech.)
Secretary Noriaki Suetake(Kyushu Inst. of Tech.) / Tomoaki Kimura(Tokyo Metropolitan Univ.)
Assistant Takanori Koga(National Inst. of Tech. Tokuyama College) / Hideaki Misawa(National Inst. of Tech., Ube College)

Paper Information
Registration To Technical Committee on Smart Info-Media Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Switching Noise Removal Filter Based on Convolutional Neural Networks and Its Application to Random-valued Impulse Noise
Sub Title (in English)
Keyword(1) Convolutional Neural Networks
Keyword(2) Denoising
Keyword(3) Switching Denoising Filter
Keyword(4) Random-valued Impulse Noise
1st Author's Name Yukiya Fukuda
1st Author's Affiliation National Institute of Technology, Ube College(NIT, UC)
2nd Author's Name Ryosuke Kubota
2nd Author's Affiliation National Institute of Technology, Ube College(NIT, UC)
Date 2018-12-07
Paper # SIS2018-29
Volume (vol) vol.118
Number (no) SIS-346
Page pp.pp.41-46(SIS),
#Pages 6
Date of Issue 2018-11-29 (SIS)