An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images
Superresolution (SR) has provided an effective solution to the increasing need for high-resolution images in remote sensing applications. Among various SR methods, deep learning-based SR (DLSR) has made a significant breakthrough. However, supervised DLSR methods require a considerable amount of tra...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9086776/ |
| _version_ | 1819105524884439040 |
|---|---|
| author | Mohammad Moein Sheikholeslami Saeed Nadi Amin Alizadeh Naeini Pedram Ghamisi |
| author_facet | Mohammad Moein Sheikholeslami Saeed Nadi Amin Alizadeh Naeini Pedram Ghamisi |
| author_sort | Mohammad Moein Sheikholeslami |
| collection | DOAJ |
| description | Superresolution (SR) has provided an effective solution to the increasing need for high-resolution images in remote sensing applications. Among various SR methods, deep learning-based SR (DLSR) has made a significant breakthrough. However, supervised DLSR methods require a considerable amount of training data, which is hardly available in the remote sensing field. To address this issue, some research works have recently proposed and revealed the capability of deep learning in unsupervised SR. This article presents an efficient unsupervised SR (EUSR) deep learning model using dense skip connections, which boosts the reconstruction performance in parallel with the reduction of computational burden. To do this, several blocks containing densely connected convolutional layers are implemented to increase the depth of the model. Some skip connections also concatenate feature maps of different blocks to enable better SR performance. Moreover, a bottle-neck block abstracts the feature maps in fewer feature maps to remarkably reduce the computational burden. According to our experiments, the proposed EUSR leads to better results than the state-of-the-art DLSR method in terms of reconstruction quality with less computational burden. Furthermore, results indicate that the EUSR is more robust than its rival in dealing with images of different classes and larger sizes. |
| first_indexed | 2024-12-22T02:23:37Z |
| format | Article |
| id | doaj.art-dde1f26b776f4891b30cc73910cd5640 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-22T02:23:37Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-dde1f26b776f4891b30cc73910cd56402022-12-21T18:42:03ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352020-01-01131937194510.1109/JSTARS.2020.29845899086776An Efficient Deep Unsupervised Superresolution Model for Remote Sensing ImagesMohammad Moein Sheikholeslami0https://orcid.org/0000-0003-4592-826XSaeed Nadi1https://orcid.org/0000-0002-7995-3714Amin Alizadeh Naeini2https://orcid.org/0000-0001-7578-6245Pedram Ghamisi3https://orcid.org/0000-0003-1203-741XDepartment of Geomatics Engineering, Faculty of Civil Engineering and Transportation, University of Isfahan, Isfahan, IranDepartment of Geomatics Engineering, Faculty of Civil Engineering and Transportation, University of Isfahan, Isfahan, IranDepartment of Geomatics Engineering, Faculty of Civil Engineering and Transportation, University of Isfahan, Isfahan, IranHelmholtz-Zentrum Dresden-Rossendorf, Helmholtz Institute Freiberg for Resource Technology, Freiberg, GermanySuperresolution (SR) has provided an effective solution to the increasing need for high-resolution images in remote sensing applications. Among various SR methods, deep learning-based SR (DLSR) has made a significant breakthrough. However, supervised DLSR methods require a considerable amount of training data, which is hardly available in the remote sensing field. To address this issue, some research works have recently proposed and revealed the capability of deep learning in unsupervised SR. This article presents an efficient unsupervised SR (EUSR) deep learning model using dense skip connections, which boosts the reconstruction performance in parallel with the reduction of computational burden. To do this, several blocks containing densely connected convolutional layers are implemented to increase the depth of the model. Some skip connections also concatenate feature maps of different blocks to enable better SR performance. Moreover, a bottle-neck block abstracts the feature maps in fewer feature maps to remarkably reduce the computational burden. According to our experiments, the proposed EUSR leads to better results than the state-of-the-art DLSR method in terms of reconstruction quality with less computational burden. Furthermore, results indicate that the EUSR is more robust than its rival in dealing with images of different classes and larger sizes.https://ieeexplore.ieee.org/document/9086776/Deep learningremote sensingsuperresolution (SR) |
| spellingShingle | Mohammad Moein Sheikholeslami Saeed Nadi Amin Alizadeh Naeini Pedram Ghamisi An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning remote sensing superresolution (SR) |
| title | An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images |
| title_full | An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images |
| title_fullStr | An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images |
| title_full_unstemmed | An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images |
| title_short | An Efficient Deep Unsupervised Superresolution Model for Remote Sensing Images |
| title_sort | efficient deep unsupervised superresolution model for remote sensing images |
| topic | Deep learning remote sensing superresolution (SR) |
| url | https://ieeexplore.ieee.org/document/9086776/ |
| work_keys_str_mv | AT mohammadmoeinsheikholeslami anefficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT saeednadi anefficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT aminalizadehnaeini anefficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT pedramghamisi anefficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT mohammadmoeinsheikholeslami efficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT saeednadi efficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT aminalizadehnaeini efficientdeepunsupervisedsuperresolutionmodelforremotesensingimages AT pedramghamisi efficientdeepunsupervisedsuperresolutionmodelforremotesensingimages |