Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution
Deep convolutional neural networks (CNNs) have been widely used and achieved state-of-the-art performance in many image or video processing and analysis tasks. In particular, for image super-resolution (SR) processing, previous CNN-based methods have led to significant improvements, when compared wi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/10/11/1700 |
| _version_ | 1828142747821801472 |
|---|---|
| author | Kui Jiang Zhongyuan Wang Peng Yi Junjun Jiang Jing Xiao Yuan Yao |
| author_facet | Kui Jiang Zhongyuan Wang Peng Yi Junjun Jiang Jing Xiao Yuan Yao |
| author_sort | Kui Jiang |
| collection | DOAJ |
| description | Deep convolutional neural networks (CNNs) have been widely used and achieved state-of-the-art performance in many image or video processing and analysis tasks. In particular, for image super-resolution (SR) processing, previous CNN-based methods have led to significant improvements, when compared with shallow learning-based methods. However, previous CNN-based algorithms with simple direct or skip connections are of poor performance when applied to remote sensing satellite images SR. In this study, a simple but effective CNN framework, namely deep distillation recursive network (DDRN), is presented for video satellite image SR. DDRN includes a group of ultra-dense residual blocks (UDB), a multi-scale purification unit (MSPU), and a reconstruction module. In particular, through the addition of rich interactive links in and between multiple-path units in each UDB, features extracted from multiple parallel convolution layers can be shared effectively. Compared with classical dense-connection-based models, DDRN possesses the following main properties. (1) DDRN contains more linking nodes with the same convolution layers. (2) A distillation and compensation mechanism, which performs feature distillation and compensation in different stages of the network, is also constructed. In particular, the high-frequency components lost during information propagation can be compensated in MSPU. (3) The final SR image can benefit from the feature maps extracted from UDB and the compensated components obtained from MSPU. Experiments on Kaggle Open Source Dataset and Jilin-1 video satellite images illustrate that DDRN outperforms the conventional CNN-based baselines and some state-of-the-art feature extraction approaches. |
| first_indexed | 2024-04-11T19:46:00Z |
| format | Article |
| id | doaj.art-60b94088a2424c6597ddaaf9c7357910 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-04-11T19:46:00Z |
| publishDate | 2018-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-60b94088a2424c6597ddaaf9c73579102022-12-22T04:06:29ZengMDPI AGRemote Sensing2072-42922018-10-011011170010.3390/rs10111700rs10111700Deep Distillation Recursive Network for Remote Sensing Imagery Super-ResolutionKui Jiang0Zhongyuan Wang1Peng Yi2Junjun Jiang3Jing Xiao4Yuan Yao5National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan 430072, ChinaNational Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan 430072, ChinaNational Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan 430072, ChinaSchool of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, ChinaNational Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan 430072, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaDeep convolutional neural networks (CNNs) have been widely used and achieved state-of-the-art performance in many image or video processing and analysis tasks. In particular, for image super-resolution (SR) processing, previous CNN-based methods have led to significant improvements, when compared with shallow learning-based methods. However, previous CNN-based algorithms with simple direct or skip connections are of poor performance when applied to remote sensing satellite images SR. In this study, a simple but effective CNN framework, namely deep distillation recursive network (DDRN), is presented for video satellite image SR. DDRN includes a group of ultra-dense residual blocks (UDB), a multi-scale purification unit (MSPU), and a reconstruction module. In particular, through the addition of rich interactive links in and between multiple-path units in each UDB, features extracted from multiple parallel convolution layers can be shared effectively. Compared with classical dense-connection-based models, DDRN possesses the following main properties. (1) DDRN contains more linking nodes with the same convolution layers. (2) A distillation and compensation mechanism, which performs feature distillation and compensation in different stages of the network, is also constructed. In particular, the high-frequency components lost during information propagation can be compensated in MSPU. (3) The final SR image can benefit from the feature maps extracted from UDB and the compensated components obtained from MSPU. Experiments on Kaggle Open Source Dataset and Jilin-1 video satellite images illustrate that DDRN outperforms the conventional CNN-based baselines and some state-of-the-art feature extraction approaches.https://www.mdpi.com/2072-4292/10/11/1700remote sensing imagerysuper-resolutionultra-dense connectionfeature distillationvideo satellitecompensation unit |
| spellingShingle | Kui Jiang Zhongyuan Wang Peng Yi Junjun Jiang Jing Xiao Yuan Yao Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution Remote Sensing remote sensing imagery super-resolution ultra-dense connection feature distillation video satellite compensation unit |
| title | Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution |
| title_full | Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution |
| title_fullStr | Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution |
| title_full_unstemmed | Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution |
| title_short | Deep Distillation Recursive Network for Remote Sensing Imagery Super-Resolution |
| title_sort | deep distillation recursive network for remote sensing imagery super resolution |
| topic | remote sensing imagery super-resolution ultra-dense connection feature distillation video satellite compensation unit |
| url | https://www.mdpi.com/2072-4292/10/11/1700 |
| work_keys_str_mv | AT kuijiang deepdistillationrecursivenetworkforremotesensingimagerysuperresolution AT zhongyuanwang deepdistillationrecursivenetworkforremotesensingimagerysuperresolution AT pengyi deepdistillationrecursivenetworkforremotesensingimagerysuperresolution AT junjunjiang deepdistillationrecursivenetworkforremotesensingimagerysuperresolution AT jingxiao deepdistillationrecursivenetworkforremotesensingimagerysuperresolution AT yuanyao deepdistillationrecursivenetworkforremotesensingimagerysuperresolution |