Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration
Autonomous underwater vehicles (AUVs) equipped with online visual inspection systems can detect underwater targets during underwater operations, which is of great significance to subsea exploration. However, the undersea scene has some instinctive challenging problems, such as poor lighting conditio...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-05-01
|
| Series: | Frontiers in Marine Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmars.2023.1112310/full |
| _version_ | 1797821322955325440 |
|---|---|
| author | Gaofei Xu Daoxian Zhou Daoxian Zhou Daoxian Zhou Libiao Yuan Libiao Yuan Libiao Yuan Wei Guo Zepeng Huang Yinlong Zhang Yinlong Zhang Yinlong Zhang |
| author_facet | Gaofei Xu Daoxian Zhou Daoxian Zhou Daoxian Zhou Libiao Yuan Libiao Yuan Libiao Yuan Wei Guo Zepeng Huang Yinlong Zhang Yinlong Zhang Yinlong Zhang |
| author_sort | Gaofei Xu |
| collection | DOAJ |
| description | Autonomous underwater vehicles (AUVs) equipped with online visual inspection systems can detect underwater targets during underwater operations, which is of great significance to subsea exploration. However, the undersea scene has some instinctive challenging problems, such as poor lighting conditions, sediment burial, and marine biofouling mimicry, which makes it difficult for traditional target detection algorithms to achieve online, reliable, and accurate detection of underwater targets. To solve the above issues, this paper proposes a real-time object detection algorithm for underwater targets based on a lightweight convolutional neural network model. To improve the imaging quality of underwater images, contrast limited adaptive histogram equalization with the fused multicolor space (FCLAHE) model is designed to enhance the image quality of underwater targets. Afterwards, a spindle-shaped backbone network is designed. The inverted residual block and group convolutions are used to extract depth features to ensure the target detection accuracy on one hand and to reduce the model parameter volume on the other hand under complex scenarios. Through extensive experiments, the precision, recall, and mAP of the proposed algorithm reached 91.2%, 90.1%, and 88.3%, respectively. It is also noticeable that the proposed method has been integrated into the embedded GPU platform and deployed in the AUV system in the practical scenarios. The average computational time is 0.053s, which satisfies the requirements of real-time object detection. |
| first_indexed | 2024-03-13T09:50:28Z |
| format | Article |
| id | doaj.art-df6cf83e7913418796ff42ee82a4c33d |
| institution | Directory Open Access Journal |
| issn | 2296-7745 |
| language | English |
| last_indexed | 2024-03-13T09:50:28Z |
| publishDate | 2023-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Marine Science |
| spelling | doaj.art-df6cf83e7913418796ff42ee82a4c33d2023-05-24T12:29:58ZengFrontiers Media S.A.Frontiers in Marine Science2296-77452023-05-011010.3389/fmars.2023.11123101112310Vision-based underwater target real-time detection for autonomous underwater vehicle subsea explorationGaofei Xu0Daoxian Zhou1Daoxian Zhou2Daoxian Zhou3Libiao Yuan4Libiao Yuan5Libiao Yuan6Wei Guo7Zepeng Huang8Yinlong Zhang9Yinlong Zhang10Yinlong Zhang11Institute of Deep-Sea Science and Engineering, Chinese Academy of Sciences, Sanya, ChinaState Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, ChinaKey Laboratory of Networked Control Systems, Chinese Academy of Sciences, Shenyang, ChinaInstitutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, ChinaState Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, ChinaKey Laboratory of Networked Control Systems, Chinese Academy of Sciences, Shenyang, ChinaInstitutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, ChinaInstitute of Deep-Sea Science and Engineering, Chinese Academy of Sciences, Sanya, ChinaUnderwater Archaeology Department, National Center for Archaeology, Beijing, ChinaState Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, ChinaKey Laboratory of Networked Control Systems, Chinese Academy of Sciences, Shenyang, ChinaInstitutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, ChinaAutonomous underwater vehicles (AUVs) equipped with online visual inspection systems can detect underwater targets during underwater operations, which is of great significance to subsea exploration. However, the undersea scene has some instinctive challenging problems, such as poor lighting conditions, sediment burial, and marine biofouling mimicry, which makes it difficult for traditional target detection algorithms to achieve online, reliable, and accurate detection of underwater targets. To solve the above issues, this paper proposes a real-time object detection algorithm for underwater targets based on a lightweight convolutional neural network model. To improve the imaging quality of underwater images, contrast limited adaptive histogram equalization with the fused multicolor space (FCLAHE) model is designed to enhance the image quality of underwater targets. Afterwards, a spindle-shaped backbone network is designed. The inverted residual block and group convolutions are used to extract depth features to ensure the target detection accuracy on one hand and to reduce the model parameter volume on the other hand under complex scenarios. Through extensive experiments, the precision, recall, and mAP of the proposed algorithm reached 91.2%, 90.1%, and 88.3%, respectively. It is also noticeable that the proposed method has been integrated into the embedded GPU platform and deployed in the AUV system in the practical scenarios. The average computational time is 0.053s, which satisfies the requirements of real-time object detection.https://www.frontiersin.org/articles/10.3389/fmars.2023.1112310/fullautonomous underwater vehiclesubsea explorationreal-time target detectionlightweight convolutional neural networkunderwater image enhancement |
| spellingShingle | Gaofei Xu Daoxian Zhou Daoxian Zhou Daoxian Zhou Libiao Yuan Libiao Yuan Libiao Yuan Wei Guo Zepeng Huang Yinlong Zhang Yinlong Zhang Yinlong Zhang Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration Frontiers in Marine Science autonomous underwater vehicle subsea exploration real-time target detection lightweight convolutional neural network underwater image enhancement |
| title | Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration |
| title_full | Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration |
| title_fullStr | Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration |
| title_full_unstemmed | Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration |
| title_short | Vision-based underwater target real-time detection for autonomous underwater vehicle subsea exploration |
| title_sort | vision based underwater target real time detection for autonomous underwater vehicle subsea exploration |
| topic | autonomous underwater vehicle subsea exploration real-time target detection lightweight convolutional neural network underwater image enhancement |
| url | https://www.frontiersin.org/articles/10.3389/fmars.2023.1112310/full |
| work_keys_str_mv | AT gaofeixu visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT daoxianzhou visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT daoxianzhou visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT daoxianzhou visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT libiaoyuan visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT libiaoyuan visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT libiaoyuan visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT weiguo visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT zepenghuang visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT yinlongzhang visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT yinlongzhang visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration AT yinlongzhang visionbasedunderwatertargetrealtimedetectionforautonomousunderwatervehiclesubseaexploration |