Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task
Bridge detection in aerial images is to determine whether a given aerial image contains one or more bridges and locate them. However, the arbitrary orientations, extreme aspect ratios, and variable backgrounds pose great challenges for bridge detection and positioning. In this article, we tackle the...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9540370/ |
| _version_ | 1819041175823187968 |
|---|---|
| author | Haowen Guo Ruixiang Zhang Yuxuan Wang Wen Yang Heng-Chao Li Gui-Song Xia |
| author_facet | Haowen Guo Ruixiang Zhang Yuxuan Wang Wen Yang Heng-Chao Li Gui-Song Xia |
| author_sort | Haowen Guo |
| collection | DOAJ |
| description | Bridge detection in aerial images is to determine whether a given aerial image contains one or more bridges and locate them. However, the arbitrary orientations, extreme aspect ratios, and variable backgrounds pose great challenges for bridge detection and positioning. In this article, we tackle these problems by combining the strengths of semantic-segmentation-based auxiliary supervision, waterbody constraint, and instance-switching-based data augmentation. More precisely, we make three main contributions. First, we propose an oriented bridge detection model with an auxiliary task of waterbody segmentation, which performs as guidance for bridge localization. The network is specifically designed in cascade style to handle the bridge detection and waterbody segmentation task end-to-end. Second, we make use of the semantic features of waterbody as spatial attention to distinguish bridges from cluttered backgrounds and then generate the waterbody segmentation map as the waterbody constraint, which introduces the prior knowledge of bridge distribution to refine the network predictions. Third, we propose a background consistent instance switching method for online data augmentation to further improve the robustness of bridge detection. To verify the effectiveness of the proposed method, we introduce a dataset named BridgeDetV1 containing 5000 well-annotated images with two kinds of bridge representations, i.e., the horizontal bounding box and the oriented bounding box. Extensive experiments demonstrate that our approach outperforms the state-of-the-art methods on this challenging benchmark. Dataset and code are available at <uri>https://github.com/whughw/BridgeDet</uri>. |
| first_indexed | 2024-12-21T09:20:49Z |
| format | Article |
| id | doaj.art-a344060426a74b29bd230f6fcc8fc8ed |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-21T09:20:49Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-a344060426a74b29bd230f6fcc8fc8ed2022-12-21T19:09:00ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01149651966610.1109/JSTARS.2021.31127059540370Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction TaskHaowen Guo0https://orcid.org/0000-0003-2328-642XRuixiang Zhang1Yuxuan Wang2Wen Yang3https://orcid.org/0000-0002-3263-8768Heng-Chao Li4https://orcid.org/0000-0002-9735-570XGui-Song Xia5https://orcid.org/0000-0001-7660-6090School of Electronic Information, Wuhan University, Wuhan, ChinaSchool of Electronic Information, Wuhan University, Wuhan, ChinaSchool of Electronic Information, Wuhan University, Wuhan, ChinaSchool of Electronic Information, Wuhan University, Wuhan, ChinaSchool of Information Science and Technology, Southwest Jiaotong University, Chengdu, ChinaSchool of Computer Science and the State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaBridge detection in aerial images is to determine whether a given aerial image contains one or more bridges and locate them. However, the arbitrary orientations, extreme aspect ratios, and variable backgrounds pose great challenges for bridge detection and positioning. In this article, we tackle these problems by combining the strengths of semantic-segmentation-based auxiliary supervision, waterbody constraint, and instance-switching-based data augmentation. More precisely, we make three main contributions. First, we propose an oriented bridge detection model with an auxiliary task of waterbody segmentation, which performs as guidance for bridge localization. The network is specifically designed in cascade style to handle the bridge detection and waterbody segmentation task end-to-end. Second, we make use of the semantic features of waterbody as spatial attention to distinguish bridges from cluttered backgrounds and then generate the waterbody segmentation map as the waterbody constraint, which introduces the prior knowledge of bridge distribution to refine the network predictions. Third, we propose a background consistent instance switching method for online data augmentation to further improve the robustness of bridge detection. To verify the effectiveness of the proposed method, we introduce a dataset named BridgeDetV1 containing 5000 well-annotated images with two kinds of bridge representations, i.e., the horizontal bounding box and the oriented bounding box. Extensive experiments demonstrate that our approach outperforms the state-of-the-art methods on this challenging benchmark. Dataset and code are available at <uri>https://github.com/whughw/BridgeDet</uri>.https://ieeexplore.ieee.org/document/9540370/Aerial imagesauxiliary taskbridge detectionconvolutional network |
| spellingShingle | Haowen Guo Ruixiang Zhang Yuxuan Wang Wen Yang Heng-Chao Li Gui-Song Xia Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Aerial images auxiliary task bridge detection convolutional network |
| title | Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task |
| title_full | Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task |
| title_fullStr | Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task |
| title_full_unstemmed | Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task |
| title_short | Accurate Bridge Detection in Aerial Images With an Auxiliary Waterbody Extraction Task |
| title_sort | accurate bridge detection in aerial images with an auxiliary waterbody extraction task |
| topic | Aerial images auxiliary task bridge detection convolutional network |
| url | https://ieeexplore.ieee.org/document/9540370/ |
| work_keys_str_mv | AT haowenguo accuratebridgedetectioninaerialimageswithanauxiliarywaterbodyextractiontask AT ruixiangzhang accuratebridgedetectioninaerialimageswithanauxiliarywaterbodyextractiontask AT yuxuanwang accuratebridgedetectioninaerialimageswithanauxiliarywaterbodyextractiontask AT wenyang accuratebridgedetectioninaerialimageswithanauxiliarywaterbodyextractiontask AT hengchaoli accuratebridgedetectioninaerialimageswithanauxiliarywaterbodyextractiontask AT guisongxia accuratebridgedetectioninaerialimageswithanauxiliarywaterbodyextractiontask |