MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection
Object detection is a critical and demanding topic in the subject of processing satellite and airborne images. The targets acquired in remote sensing imagery are at various sizes, and the backgrounds are complicated, which makes object detection extremely challenging. We address these aforementioned...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10064169/ |
| _version_ | 1797859061482389504 |
|---|---|
| author | Keyan Wang Feiyu Bai Jiaojiao Li Yajing Liu Yunsong Li |
| author_facet | Keyan Wang Feiyu Bai Jiaojiao Li Yajing Liu Yunsong Li |
| author_sort | Keyan Wang |
| collection | DOAJ |
| description | Object detection is a critical and demanding topic in the subject of processing satellite and airborne images. The targets acquired in remote sensing imagery are at various sizes, and the backgrounds are complicated, which makes object detection extremely challenging. We address these aforementioned issues in this article by introducing the MashFormer, an innovative multiscale aware convolutional neural network (CNN) and transformer integrated hybrid detector. Specifically, MashFormer employs the transformer block to complement the CNN-based feature extraction backbone, which could obtain the relationships between long-range features and enhance the representative ability in complex background scenarios. With the intention of improving the detection performance for objects with multiscale characteristic, since in remote sensing scenarios, the size of object varies greatly. A multilevel feature aggregation component, incorporate with a cross-level feature alignment module is designed to alleviate the semantic discrepancy between features from shallow and deep layers. To verify the effectiveness of the suggested MashFormer, comparative experiments are carried out with other cutting-edge methodologies using the publicly available high resolution remote sensing detection and Northwestern Polytechnical University VHR-10 datasets. The experimental findings confirm the effectiveness and superiority of our suggested model by indicating that our approach has greater mean average precision than the other methodologies. |
| first_indexed | 2024-04-09T21:23:24Z |
| format | Article |
| id | doaj.art-0c9fa8148949498aaaf4eed7557b7346 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-04-09T21:23:24Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-0c9fa8148949498aaaf4eed7557b73462023-03-27T23:00:05ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352023-01-01162753276310.1109/JSTARS.2023.325404710064169MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object DetectionKeyan Wang0https://orcid.org/0000-0002-9545-718XFeiyu Bai1https://orcid.org/0009-0002-4180-7954Jiaojiao Li2https://orcid.org/0000-0002-0470-9469Yajing Liu3Yunsong Li4https://orcid.org/0000-0002-0234-6270State Key Laboratory of Integrated Service Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaState Key Laboratory of Integrated Service Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaState Key Laboratory of Integrated Service Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaState Key Laboratory of Integrated Service Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaState Key Laboratory of Integrated Service Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaObject detection is a critical and demanding topic in the subject of processing satellite and airborne images. The targets acquired in remote sensing imagery are at various sizes, and the backgrounds are complicated, which makes object detection extremely challenging. We address these aforementioned issues in this article by introducing the MashFormer, an innovative multiscale aware convolutional neural network (CNN) and transformer integrated hybrid detector. Specifically, MashFormer employs the transformer block to complement the CNN-based feature extraction backbone, which could obtain the relationships between long-range features and enhance the representative ability in complex background scenarios. With the intention of improving the detection performance for objects with multiscale characteristic, since in remote sensing scenarios, the size of object varies greatly. A multilevel feature aggregation component, incorporate with a cross-level feature alignment module is designed to alleviate the semantic discrepancy between features from shallow and deep layers. To verify the effectiveness of the suggested MashFormer, comparative experiments are carried out with other cutting-edge methodologies using the publicly available high resolution remote sensing detection and Northwestern Polytechnical University VHR-10 datasets. The experimental findings confirm the effectiveness and superiority of our suggested model by indicating that our approach has greater mean average precision than the other methodologies.https://ieeexplore.ieee.org/document/10064169/Feature alignmentobject detectionremote sensingtransformer |
| spellingShingle | Keyan Wang Feiyu Bai Jiaojiao Li Yajing Liu Yunsong Li MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Feature alignment object detection remote sensing transformer |
| title | MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection |
| title_full | MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection |
| title_fullStr | MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection |
| title_full_unstemmed | MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection |
| title_short | MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object Detection |
| title_sort | mashformer a novel multiscale aware hybrid detector for remote sensing object detection |
| topic | Feature alignment object detection remote sensing transformer |
| url | https://ieeexplore.ieee.org/document/10064169/ |
| work_keys_str_mv | AT keyanwang mashformeranovelmultiscaleawarehybriddetectorforremotesensingobjectdetection AT feiyubai mashformeranovelmultiscaleawarehybriddetectorforremotesensingobjectdetection AT jiaojiaoli mashformeranovelmultiscaleawarehybriddetectorforremotesensingobjectdetection AT yajingliu mashformeranovelmultiscaleawarehybriddetectorforremotesensingobjectdetection AT yunsongli mashformeranovelmultiscaleawarehybriddetectorforremotesensingobjectdetection |