Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution
Remote sensing images are essential in many fields, such as land cover classification and building extraction. The huge difference between the directly acquired remote sensing images and the actual scene, due to the complex degradation process and hardware limitations, seriously affects the performa...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10319071/ |
| _version_ | 1797376438608855040 |
|---|---|
| author | Zhihan Ren Lijun He Jichuan Lu |
| author_facet | Zhihan Ren Lijun He Jichuan Lu |
| author_sort | Zhihan Ren |
| collection | DOAJ |
| description | Remote sensing images are essential in many fields, such as land cover classification and building extraction. The huge difference between the directly acquired remote sensing images and the actual scene, due to the complex degradation process and hardware limitations, seriously affects the performance achieved by the same classification or segmentation model. Therefore, using super-resolution (SR) algorithms to improve image quality and achieve better results is an effective method. However, current SR methods only focus on the similarity of pixel values between SR and high-resolution (HR) images without considering perceptual similarities, which usually leads to the problem of oversmoothed and blurred edge details. Moreover, there is little attention to human visual habits and machine vision applications for remote sensing images. In this work, we propose the context aware edge-enhanced generative adversarial network (CEEGAN) SR framework to reconstruct visually pleasing images that can be practically applied in actual scenarios. In the generator of CEEGAN, we build an edge feature enhanced module (EFEM) to enhance the edges by combining the edge features with context information. Edge restoration block is designed to fuse multiscale edge features enhanced by EFEM and reconstruct a refined edge map. Furthermore, we designed an edge loss function to constrain the generated SR and HR similarity at the edge domain. Experimental results show that our proposed method can obtain SR images with a better reconstruction performance. Meanwhile, CEEGAN can achieve the best results on classification and semantic segmentation datasets for machine vision applications. |
| first_indexed | 2024-03-08T19:38:42Z |
| format | Article |
| id | doaj.art-a837e469932c4470a5ed867b45ebf5bf |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-08T19:38:42Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-a837e469932c4470a5ed867b45ebf5bf2023-12-26T00:01:04ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01171363137610.1109/JSTARS.2023.333327110319071Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-ResolutionZhihan Ren0https://orcid.org/0009-0003-5763-4581Lijun He1https://orcid.org/0000-0002-3911-8263Jichuan Lu2https://orcid.org/0009-0008-2162-971XShaanxi Key Laboratory of Deep Space Exploration Intelligent Information Technology, School of Information and Communications Engineering, Xi'an Jiaotong University, Xi'an, ChinaShaanxi Key Laboratory of Deep Space Exploration Intelligent Information Technology, School of Information and Communications Engineering, Xi'an Jiaotong University, Xi'an, ChinaChina Mobile Communications Group Shaanxi Company Ltd., Xi'an, ChinaRemote sensing images are essential in many fields, such as land cover classification and building extraction. The huge difference between the directly acquired remote sensing images and the actual scene, due to the complex degradation process and hardware limitations, seriously affects the performance achieved by the same classification or segmentation model. Therefore, using super-resolution (SR) algorithms to improve image quality and achieve better results is an effective method. However, current SR methods only focus on the similarity of pixel values between SR and high-resolution (HR) images without considering perceptual similarities, which usually leads to the problem of oversmoothed and blurred edge details. Moreover, there is little attention to human visual habits and machine vision applications for remote sensing images. In this work, we propose the context aware edge-enhanced generative adversarial network (CEEGAN) SR framework to reconstruct visually pleasing images that can be practically applied in actual scenarios. In the generator of CEEGAN, we build an edge feature enhanced module (EFEM) to enhance the edges by combining the edge features with context information. Edge restoration block is designed to fuse multiscale edge features enhanced by EFEM and reconstruct a refined edge map. Furthermore, we designed an edge loss function to constrain the generated SR and HR similarity at the edge domain. Experimental results show that our proposed method can obtain SR images with a better reconstruction performance. Meanwhile, CEEGAN can achieve the best results on classification and semantic segmentation datasets for machine vision applications.https://ieeexplore.ieee.org/document/10319071/Edge enhancementgenerative adversarial network (GAN)remote sensing imagessuper-resolution (SR) |
| spellingShingle | Zhihan Ren Lijun He Jichuan Lu Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Edge enhancement generative adversarial network (GAN) remote sensing images super-resolution (SR) |
| title | Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution |
| title_full | Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution |
| title_fullStr | Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution |
| title_full_unstemmed | Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution |
| title_short | Context Aware Edge-Enhanced GAN for Remote Sensing Image Super-Resolution |
| title_sort | context aware edge enhanced gan for remote sensing image super resolution |
| topic | Edge enhancement generative adversarial network (GAN) remote sensing images super-resolution (SR) |
| url | https://ieeexplore.ieee.org/document/10319071/ |
| work_keys_str_mv | AT zhihanren contextawareedgeenhancedganforremotesensingimagesuperresolution AT lijunhe contextawareedgeenhancedganforremotesensingimagesuperresolution AT jichuanlu contextawareedgeenhancedganforremotesensingimagesuperresolution |