Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy
Current synthetic aperture radar (SAR) ship datasets exhibit a notable disparity in the distribution of large, medium, and small ship targets. This imbalance makes it difficult for a relatively small number of large and medium-sized ships to be effectively trained, resulting in many false alarms. Th...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/13/2257 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849704673017069568 |
|---|---|
| author | Shun He Ruirui Yuan Zhiwei Yang Jiaxue Liu |
| author_facet | Shun He Ruirui Yuan Zhiwei Yang Jiaxue Liu |
| author_sort | Shun He |
| collection | DOAJ |
| description | Current synthetic aperture radar (SAR) ship datasets exhibit a notable disparity in the distribution of large, medium, and small ship targets. This imbalance makes it difficult for a relatively small number of large and medium-sized ships to be effectively trained, resulting in many false alarms. Therefore, to address the issues of scale diversity, intra-class imbalance in ship data, and the feature conflict problem associated with traditional coupled detection heads, we propose an SAR image multiscale task-decoupled oriented ship target detector based on a size-aware balanced strategy. First, the multiscale target features are extracted using the multikernel heterogeneous perception module (MKHP). Meanwhile, the triple-attention module is introduced to establish the remote channel dependence to alleviate the issue of small target feature annihilation, which can effectively enhance the feature characterization ability of the model. Second, given the differences in the demand for feature information between the detection and classification tasks, a channel attention-based task decoupling dual-head (CAT2D) detector head structure is introduced to address the inherent conflict between classification and localization tasks. Finally, a new size-aware balanced (SAB) loss strategy is proposed to guide the network in focusing on the scarce targets in training to alleviate the intra-class imbalance problem during the training process. The ablation experiments on SSDD+ reflect the contribution of each component, and the results of the comparison experiments on the RSDD-SAR and HRSID datasets show that the proposed method achieves state-of-the-art performance compared to other state-of-the-art detection models. Furthermore, our approach exhibits superior detection coverage for both offshore and inshore scenarios for ship detection tasks. |
| format | Article |
| id | doaj-art-0d60dc0dc4de4b7eab46e477739f657a |
| institution | DOAJ |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-0d60dc0dc4de4b7eab46e477739f657a2025-08-20T03:16:42ZengMDPI AGRemote Sensing2072-42922025-06-011713225710.3390/rs17132257Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced StrategyShun He0Ruirui Yuan1Zhiwei Yang2Jiaxue Liu3School of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710600, ChinaSchool of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710600, ChinaSchool of Electronic Engineering, Xidian University, Xi’an 710126, ChinaSchool of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710600, ChinaCurrent synthetic aperture radar (SAR) ship datasets exhibit a notable disparity in the distribution of large, medium, and small ship targets. This imbalance makes it difficult for a relatively small number of large and medium-sized ships to be effectively trained, resulting in many false alarms. Therefore, to address the issues of scale diversity, intra-class imbalance in ship data, and the feature conflict problem associated with traditional coupled detection heads, we propose an SAR image multiscale task-decoupled oriented ship target detector based on a size-aware balanced strategy. First, the multiscale target features are extracted using the multikernel heterogeneous perception module (MKHP). Meanwhile, the triple-attention module is introduced to establish the remote channel dependence to alleviate the issue of small target feature annihilation, which can effectively enhance the feature characterization ability of the model. Second, given the differences in the demand for feature information between the detection and classification tasks, a channel attention-based task decoupling dual-head (CAT2D) detector head structure is introduced to address the inherent conflict between classification and localization tasks. Finally, a new size-aware balanced (SAB) loss strategy is proposed to guide the network in focusing on the scarce targets in training to alleviate the intra-class imbalance problem during the training process. The ablation experiments on SSDD+ reflect the contribution of each component, and the results of the comparison experiments on the RSDD-SAR and HRSID datasets show that the proposed method achieves state-of-the-art performance compared to other state-of-the-art detection models. Furthermore, our approach exhibits superior detection coverage for both offshore and inshore scenarios for ship detection tasks.https://www.mdpi.com/2072-4292/17/13/2257synthetic aperture radar (SAR)oriented ship detectionintra-class imbalancemulti-scale detectiontask decoupling |
| spellingShingle | Shun He Ruirui Yuan Zhiwei Yang Jiaxue Liu Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy Remote Sensing synthetic aperture radar (SAR) oriented ship detection intra-class imbalance multi-scale detection task decoupling |
| title | Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy |
| title_full | Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy |
| title_fullStr | Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy |
| title_full_unstemmed | Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy |
| title_short | Multiscale Task-Decoupled Oriented SAR Ship Detection Network Based on Size-Aware Balanced Strategy |
| title_sort | multiscale task decoupled oriented sar ship detection network based on size aware balanced strategy |
| topic | synthetic aperture radar (SAR) oriented ship detection intra-class imbalance multi-scale detection task decoupling |
| url | https://www.mdpi.com/2072-4292/17/13/2257 |
| work_keys_str_mv | AT shunhe multiscaletaskdecoupledorientedsarshipdetectionnetworkbasedonsizeawarebalancedstrategy AT ruiruiyuan multiscaletaskdecoupledorientedsarshipdetectionnetworkbasedonsizeawarebalancedstrategy AT zhiweiyang multiscaletaskdecoupledorientedsarshipdetectionnetworkbasedonsizeawarebalancedstrategy AT jiaxueliu multiscaletaskdecoupledorientedsarshipdetectionnetworkbasedonsizeawarebalancedstrategy |