Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN

Removing snow particles from an image is a complicated task due to the particles’ shape, size, and color. The latest snow removal methods remove snow from a single image but retain some snow and salt-and-pepper particles. Some approaches, while trying to remove snow from a single image, p...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Thaileang Sung, Hyo Jong Lee
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Snow removal de-snow deep learning residual network convolutional neural network discrete wavelet transformation Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/16014842dee649498f797e4a289e5338
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:16014842dee649498f797e4a289e5338
record_format	dspace
spelling	oai:doaj.org-article:16014842dee649498f797e4a289e53382021-11-18T00:01:09ZRemoving Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN2169-353610.1109/ACCESS.2021.3126539https://doaj.org/article/16014842dee649498f797e4a289e53382021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9606733/https://doaj.org/toc/2169-3536Removing snow particles from an image is a complicated task due to the particles’ shape, size, and color. The latest snow removal methods remove snow from a single image but retain some snow and salt-and-pepper particles. Some approaches, while trying to remove snow from a single image, produce blurry artifacts. In this paper, we solve these problems by designing a network model that consists of a residual generative network, a snow-free image generative network, and a perceptual relativistic discriminative network. In both generative networks, we assign the residual frequency network (ReFNet) as our bottleneck module. Our network model learns to map two relationships. First, the input snowy image is trained to map the snow mask image in the dataset. Then, a retained image resulting from subtraction between an input image and the estimated residual image is concatenated with the input snowy image and mapped to the desired snow-free ground truth. Moreover, we use a perceptual identical-paired adversarial network based on a relativistic discriminative network to make our training results more robust. Our results achieve greater performance than state-of-the-art methods on both synthetic and real-world snowy images.Thaileang SungHyo Jong LeeIEEEarticleSnow removalde-snowdeep learningresidual networkconvolutional neural networkdiscrete wavelet transformationElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 152047-152056 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Snow removal de-snow deep learning residual network convolutional neural network discrete wavelet transformation Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Snow removal de-snow deep learning residual network convolutional neural network discrete wavelet transformation Electrical engineering. Electronics. Nuclear engineering TK1-9971 Thaileang Sung Hyo Jong Lee Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN
description	Removing snow particles from an image is a complicated task due to the particles’ shape, size, and color. The latest snow removal methods remove snow from a single image but retain some snow and salt-and-pepper particles. Some approaches, while trying to remove snow from a single image, produce blurry artifacts. In this paper, we solve these problems by designing a network model that consists of a residual generative network, a snow-free image generative network, and a perceptual relativistic discriminative network. In both generative networks, we assign the residual frequency network (ReFNet) as our bottleneck module. Our network model learns to map two relationships. First, the input snowy image is trained to map the snow mask image in the dataset. Then, a retained image resulting from subtraction between an input image and the estimated residual image is concatenated with the input snowy image and mapped to the desired snow-free ground truth. Moreover, we use a perceptual identical-paired adversarial network based on a relativistic discriminative network to make our training results more robust. Our results achieve greater performance than state-of-the-art methods on both synthetic and real-world snowy images.
format	article
author	Thaileang Sung Hyo Jong Lee
author_facet	Thaileang Sung Hyo Jong Lee
author_sort	Thaileang Sung
title	Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN
title_short	Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN
title_full	Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN
title_fullStr	Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN
title_full_unstemmed	Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN
title_sort	removing snow from a single image using a residual frequency module and perceptual ralsgan
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/16014842dee649498f797e4a289e5338
work_keys_str_mv	AT thaileangsung removingsnowfromasingleimageusingaresidualfrequencymoduleandperceptualralsgan AT hyojonglee removingsnowfromasingleimageusingaresidualfrequencymoduleandperceptualralsgan
_version_	1718425208649416704

Removing Snow From a Single Image Using a Residual Frequency Module and Perceptual RaLSGAN

Ejemplares similares