Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images
<b> </b>Improving the resolution of degraded radar echo images of weather radar systems can aid severe weather forecasting and disaster prevention. Previous approaches to this problem include classical super-resolution (SR) algorithms such as iterative back-projection (IBP) and a recent...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-09-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/10/9/555 |
| _version_ | 1819056867506126848 |
|---|---|
| author | Hongguang Chen Xing Zhang Yintian Liu Qiangyu Zeng |
| author_facet | Hongguang Chen Xing Zhang Yintian Liu Qiangyu Zeng |
| author_sort | Hongguang Chen |
| collection | DOAJ |
| description | <b> </b>Improving the resolution of degraded radar echo images of weather radar systems can aid severe weather forecasting and disaster prevention. Previous approaches to this problem include classical super-resolution (SR) algorithms such as iterative back-projection (IBP) and a recent nonlocal self-similarity sparse representation (NSSR) that exploits the data redundancy of radar echo data, etc. However, since radar echoes tend to have rich edge information and contour textures, the textural detail in the reconstructed echoes of traditional approaches is typically absent. Inspired by the recent advances of faster and deeper neural networks, especially the generative adversarial networks (GAN), which are capable of pushing SR solutions to the natural image manifold, we propose using GAN to tackle the problem of weather radar echo super-resolution to achieve better reconstruction performance (measured in peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM)). Using authentic weather radar echo data, we present the experimental results and compare its reconstruction performance with the above-mentioned methods. The experimental results showed that the GAN-based method is capable of generating perceptually superior solutions while achieving higher PSNR/SSIM results. |
| first_indexed | 2024-12-21T13:30:14Z |
| format | Article |
| id | doaj.art-a3f50f003e07459b87e7bed69a9be84a |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-12-21T13:30:14Z |
| publishDate | 2019-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-a3f50f003e07459b87e7bed69a9be84a2022-12-21T19:02:19ZengMDPI AGAtmosphere2073-44332019-09-0110955510.3390/atmos10090555atmos10090555Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo ImagesHongguang Chen0Xing Zhang1Yintian Liu2Qiangyu Zeng3College of Software Engineering, Chengdu University of Information Technology, Chengdu 610225, ChinaCollege of Electronic Engineering, Chengdu University of Information Technology, Chengdu 610225, ChinaCollege of Software Engineering, Chengdu University of Information Technology, Chengdu 610225, ChinaCollege of Electronic Engineering, Chengdu University of Information Technology, Chengdu 610225, China<b> </b>Improving the resolution of degraded radar echo images of weather radar systems can aid severe weather forecasting and disaster prevention. Previous approaches to this problem include classical super-resolution (SR) algorithms such as iterative back-projection (IBP) and a recent nonlocal self-similarity sparse representation (NSSR) that exploits the data redundancy of radar echo data, etc. However, since radar echoes tend to have rich edge information and contour textures, the textural detail in the reconstructed echoes of traditional approaches is typically absent. Inspired by the recent advances of faster and deeper neural networks, especially the generative adversarial networks (GAN), which are capable of pushing SR solutions to the natural image manifold, we propose using GAN to tackle the problem of weather radar echo super-resolution to achieve better reconstruction performance (measured in peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM)). Using authentic weather radar echo data, we present the experimental results and compare its reconstruction performance with the above-mentioned methods. The experimental results showed that the GAN-based method is capable of generating perceptually superior solutions while achieving higher PSNR/SSIM results.https://www.mdpi.com/2073-4433/10/9/555weather radarimage super-resolutiongenerative adversarial networksdeep learning |
| spellingShingle | Hongguang Chen Xing Zhang Yintian Liu Qiangyu Zeng Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images Atmosphere weather radar image super-resolution generative adversarial networks deep learning |
| title | Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images |
| title_full | Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images |
| title_fullStr | Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images |
| title_full_unstemmed | Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images |
| title_short | Generative Adversarial Networks Capabilities for Super-Resolution Reconstruction of Weather Radar Echo Images |
| title_sort | generative adversarial networks capabilities for super resolution reconstruction of weather radar echo images |
| topic | weather radar image super-resolution generative adversarial networks deep learning |
| url | https://www.mdpi.com/2073-4433/10/9/555 |
| work_keys_str_mv | AT hongguangchen generativeadversarialnetworkscapabilitiesforsuperresolutionreconstructionofweatherradarechoimages AT xingzhang generativeadversarialnetworkscapabilitiesforsuperresolutionreconstructionofweatherradarechoimages AT yintianliu generativeadversarialnetworkscapabilitiesforsuperresolutionreconstructionofweatherradarechoimages AT qiangyuzeng generativeadversarialnetworkscapabilitiesforsuperresolutionreconstructionofweatherradarechoimages |