Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis

Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis

Ground-based weather radar can observe a wide range with a high spatial and temporal resolution. They are beneficial to meteorological research and services by providing valuable information. Recent weather radar data related research has focused on applying machine learning and deep learning to sol...

Full description

Bibliographic Details
Main Authors: Hansoo Lee, Jonggeun Kim, Eun Kyeong Kim, Sungshin Kim
Format: Article
Language:English
Published: MDPI AG 2020-02-01
Series:Applied Sciences
Subjects:
weather radar
data augmentation
generative adversarial networks
structural similarity
Online Access:https://www.mdpi.com/2076-3417/10/4/1449
id doaj-e6f1051226884c0990713403dafdc9f7
record_format Article
spelling doaj-e6f1051226884c0990713403dafdc9f72020-11-25T00:31:11ZengMDPI AGApplied Sciences2076-34172020-02-01104144910.3390/app10041449app10041449Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data AnalysisHansoo Lee0Jonggeun Kim1Eun Kyeong Kim2Sungshin Kim3Department of Electrical and Computer Engineering, Pusan National University, Busan 46241, KoreaDepartment of Electrical and Computer Engineering, Pusan National University, Busan 46241, KoreaDepartment of Electrical and Computer Engineering, Pusan National University, Busan 46241, KoreaDepartment of Electrical and Computer Engineering, Pusan National University, Busan 46241, KoreaGround-based weather radar can observe a wide range with a high spatial and temporal resolution. They are beneficial to meteorological research and services by providing valuable information. Recent weather radar data related research has focused on applying machine learning and deep learning to solve complicated problems. It is a well-known fact that an adequate amount of data is a positively necessary condition in machine learning and deep learning. Generative adversarial networks (GANs) have received extensive attention for their remarkable data generation capacity, with a fascinating competitive structure having been proposed since. Consequently, a massive number of variants have been proposed; which model is adequate to solve the given problem is an inevitable concern. In this paper, we propose exploring the problem of radar image synthesis and evaluating different GANs with authentic radar observation results. The experimental results showed that the improved Wasserstein GAN is more capable of generating similar radar images while achieving higher structural similarity results.https://www.mdpi.com/2076-3417/10/4/1449weather radardata augmentationgenerative adversarial networksstructural similarity
collection DOAJ
language English
format Article
sources DOAJ
author Hansoo Lee
Jonggeun Kim
Eun Kyeong Kim
Sungshin Kim
spellingShingle Hansoo Lee
Jonggeun Kim
Eun Kyeong Kim
Sungshin Kim
Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis
Applied Sciences
weather radar
data augmentation
generative adversarial networks
structural similarity
author_facet Hansoo Lee
Jonggeun Kim
Eun Kyeong Kim
Sungshin Kim
author_sort Hansoo Lee
title Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis
title_short Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis
title_full Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis
title_fullStr Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis
title_full_unstemmed Wasserstein Generative Adversarial Networks Based Data Augmentation for Radar Data Analysis
title_sort wasserstein generative adversarial networks based data augmentation for radar data analysis
publisher MDPI AG
series Applied Sciences
issn 2076-3417
publishDate 2020-02-01
description Ground-based weather radar can observe a wide range with a high spatial and temporal resolution. They are beneficial to meteorological research and services by providing valuable information. Recent weather radar data related research has focused on applying machine learning and deep learning to solve complicated problems. It is a well-known fact that an adequate amount of data is a positively necessary condition in machine learning and deep learning. Generative adversarial networks (GANs) have received extensive attention for their remarkable data generation capacity, with a fascinating competitive structure having been proposed since. Consequently, a massive number of variants have been proposed; which model is adequate to solve the given problem is an inevitable concern. In this paper, we propose exploring the problem of radar image synthesis and evaluating different GANs with authentic radar observation results. The experimental results showed that the improved Wasserstein GAN is more capable of generating similar radar images while achieving higher structural similarity results.
topic weather radar
data augmentation
generative adversarial networks
structural similarity
url https://www.mdpi.com/2076-3417/10/4/1449
work_keys_str_mv AT hansoolee wassersteingenerativeadversarialnetworksbaseddataaugmentationforradardataanalysis
AT jonggeunkim wassersteingenerativeadversarialnetworksbaseddataaugmentationforradardataanalysis
AT eunkyeongkim wassersteingenerativeadversarialnetworksbaseddataaugmentationforradardataanalysis
AT sungshinkim wassersteingenerativeadversarialnetworksbaseddataaugmentationforradardataanalysis
_version_ 1725323274900996096

Similar Items