Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network
In many domestic and military applications, aerial vehicle detection and super-resolution algorithms are frequently developed and applied independently. However, aerial vehicle detection on super-resolved images remains a challenging task due to the lack of discriminative information in the super-re...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9079855/ |
| _version_ | 1818929828451057664 |
|---|---|
| author | Moktari Mostofa Syeda Nyma Ferdous Benjamin S. Riggan Nasser M. Nasrabadi |
| author_facet | Moktari Mostofa Syeda Nyma Ferdous Benjamin S. Riggan Nasser M. Nasrabadi |
| author_sort | Moktari Mostofa |
| collection | DOAJ |
| description | In many domestic and military applications, aerial vehicle detection and super-resolution algorithms are frequently developed and applied independently. However, aerial vehicle detection on super-resolved images remains a challenging task due to the lack of discriminative information in the super-resolved images. To address this problem, we propose a Joint Super-Resolution and Vehicle Detection Network (Joint-SRVDNet) that tries to generate discriminative, high-resolution images of vehicles from low-resolution aerial images. First, aerial images are up-scaled by a factor of 4x using a Multi-scale Generative Adversarial Network (MsGAN), which has multiple intermediate outputs with increasing resolutions. Second, a detector is trained on super-resolved images that are upscaled by factor 4x using MsGAN architecture and finally, the detection loss is minimized jointly with the super-resolution loss to encourage the target detector to be sensitive to the subsequent super-resolution training. The network jointly learns hierarchical and discriminative features of targets and produces optimal super-resolution results. We perform both quantitative and qualitative evaluation of our proposed network on VEDAI, xView and DOTA datasets. The experimental results show that our proposed framework achieves better visual quality than the state-of-the-art methods for aerial super-resolution with 4x up-scaling factor and improves the accuracy of aerial vehicle detection. |
| first_indexed | 2024-12-20T03:51:00Z |
| format | Article |
| id | doaj.art-06924abfbabb41f18e615c89bf488dad |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T03:51:00Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-06924abfbabb41f18e615c89bf488dad2022-12-21T19:54:27ZengIEEEIEEE Access2169-35362020-01-018823068231910.1109/ACCESS.2020.29908709079855Joint-SRVDNet: Joint Super Resolution and Vehicle Detection NetworkMoktari Mostofa0https://orcid.org/0000-0001-8719-3244Syeda Nyma Ferdous1Benjamin S. Riggan2Nasser M. Nasrabadi3Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USALane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USADepartment of Electrical and Computer Engineering, University of Nebraska–Lincoln, Lincoln, NE, USALane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USAIn many domestic and military applications, aerial vehicle detection and super-resolution algorithms are frequently developed and applied independently. However, aerial vehicle detection on super-resolved images remains a challenging task due to the lack of discriminative information in the super-resolved images. To address this problem, we propose a Joint Super-Resolution and Vehicle Detection Network (Joint-SRVDNet) that tries to generate discriminative, high-resolution images of vehicles from low-resolution aerial images. First, aerial images are up-scaled by a factor of 4x using a Multi-scale Generative Adversarial Network (MsGAN), which has multiple intermediate outputs with increasing resolutions. Second, a detector is trained on super-resolved images that are upscaled by factor 4x using MsGAN architecture and finally, the detection loss is minimized jointly with the super-resolution loss to encourage the target detector to be sensitive to the subsequent super-resolution training. The network jointly learns hierarchical and discriminative features of targets and produces optimal super-resolution results. We perform both quantitative and qualitative evaluation of our proposed network on VEDAI, xView and DOTA datasets. The experimental results show that our proposed framework achieves better visual quality than the state-of-the-art methods for aerial super-resolution with 4x up-scaling factor and improves the accuracy of aerial vehicle detection.https://ieeexplore.ieee.org/document/9079855/Aerial imagesmulti-scale generative adversarial network (MsGAN)super-resolutionvehicle detection |
| spellingShingle | Moktari Mostofa Syeda Nyma Ferdous Benjamin S. Riggan Nasser M. Nasrabadi Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network IEEE Access Aerial images multi-scale generative adversarial network (MsGAN) super-resolution vehicle detection |
| title | Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network |
| title_full | Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network |
| title_fullStr | Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network |
| title_full_unstemmed | Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network |
| title_short | Joint-SRVDNet: Joint Super Resolution and Vehicle Detection Network |
| title_sort | joint srvdnet joint super resolution and vehicle detection network |
| topic | Aerial images multi-scale generative adversarial network (MsGAN) super-resolution vehicle detection |
| url | https://ieeexplore.ieee.org/document/9079855/ |
| work_keys_str_mv | AT moktarimostofa jointsrvdnetjointsuperresolutionandvehicledetectionnetwork AT syedanymaferdous jointsrvdnetjointsuperresolutionandvehicledetectionnetwork AT benjaminsriggan jointsrvdnetjointsuperresolutionandvehicledetectionnetwork AT nassermnasrabadi jointsrvdnetjointsuperresolutionandvehicledetectionnetwork |