SR-DETR: Target Detection in Maritime Rescue from UAV Imagery
The growth of maritime transportation has been accompanied by a gradual increase in accident rates, drawing greater attention to the critical issue of man-overboard incidents and drowning. Traditional maritime search-and-rescue (SAR) methods are often constrained by limited efficiency and high opera...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/12/2026 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839652754640863232 |
|---|---|
| author | Yuling Liu Yan Wei |
| author_facet | Yuling Liu Yan Wei |
| author_sort | Yuling Liu |
| collection | DOAJ |
| description | The growth of maritime transportation has been accompanied by a gradual increase in accident rates, drawing greater attention to the critical issue of man-overboard incidents and drowning. Traditional maritime search-and-rescue (SAR) methods are often constrained by limited efficiency and high operational costs. Over the past few years, drones have demonstrated significant promise in improving the effectiveness of search-and-rescue operations. This is largely due to their exceptional ability to move freely and their capacity for wide-area monitoring. This study proposes an enhanced SR-DETR algorithm aimed at improving the detection of individuals who have fallen overboard. Specifically, the conventional multi-head self-attention (MHSA) mechanism is replaced with Efficient Additive Attention (EAA), which facilitates more efficient feature interaction while substantially reducing computational complexity. Moreover, we introduce a new feature aggregation module called the Cross-Stage Partial Parallel Atrous Feature Pyramid Network (CPAFPN). By refining spatial attention mechanisms, the module significantly boosts cross-scale target recognition capabilities in the model, especially offering advantages for detecting smaller objects. To improve localization precision, we develop a novel loss function for bounding box regression, named Focaler-GIoU, which performs particularly well when handling densely packed and small-scale objects. The proposed approach is validated through experiments and achieves an mAP of 86.5%, which surpasses the baseline RT-DETR model’s performance of 83.2%. These outcomes highlight the practicality and reliability of our method in detecting individuals overboard, contributing to more precise and resource-efficient solutions for real-time maritime rescue efforts. |
| format | Article |
| id | doaj-art-16f8d24f4e154a92b814343c9ee7d8c9 |
| institution | Matheson Library |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-16f8d24f4e154a92b814343c9ee7d8c92025-06-25T14:23:32ZengMDPI AGRemote Sensing2072-42922025-06-011712202610.3390/rs17122026SR-DETR: Target Detection in Maritime Rescue from UAV ImageryYuling Liu0Yan Wei1College of Computer and Information Science, Chongqing Normal University, Chongqing 401331, ChinaCollege of Computer and Information Science, Chongqing Normal University, Chongqing 401331, ChinaThe growth of maritime transportation has been accompanied by a gradual increase in accident rates, drawing greater attention to the critical issue of man-overboard incidents and drowning. Traditional maritime search-and-rescue (SAR) methods are often constrained by limited efficiency and high operational costs. Over the past few years, drones have demonstrated significant promise in improving the effectiveness of search-and-rescue operations. This is largely due to their exceptional ability to move freely and their capacity for wide-area monitoring. This study proposes an enhanced SR-DETR algorithm aimed at improving the detection of individuals who have fallen overboard. Specifically, the conventional multi-head self-attention (MHSA) mechanism is replaced with Efficient Additive Attention (EAA), which facilitates more efficient feature interaction while substantially reducing computational complexity. Moreover, we introduce a new feature aggregation module called the Cross-Stage Partial Parallel Atrous Feature Pyramid Network (CPAFPN). By refining spatial attention mechanisms, the module significantly boosts cross-scale target recognition capabilities in the model, especially offering advantages for detecting smaller objects. To improve localization precision, we develop a novel loss function for bounding box regression, named Focaler-GIoU, which performs particularly well when handling densely packed and small-scale objects. The proposed approach is validated through experiments and achieves an mAP of 86.5%, which surpasses the baseline RT-DETR model’s performance of 83.2%. These outcomes highlight the practicality and reliability of our method in detecting individuals overboard, contributing to more precise and resource-efficient solutions for real-time maritime rescue efforts.https://www.mdpi.com/2072-4292/17/12/2026target detectionmaritime search and rescuedronescomputer vision |
| spellingShingle | Yuling Liu Yan Wei SR-DETR: Target Detection in Maritime Rescue from UAV Imagery Remote Sensing target detection maritime search and rescue drones computer vision |
| title | SR-DETR: Target Detection in Maritime Rescue from UAV Imagery |
| title_full | SR-DETR: Target Detection in Maritime Rescue from UAV Imagery |
| title_fullStr | SR-DETR: Target Detection in Maritime Rescue from UAV Imagery |
| title_full_unstemmed | SR-DETR: Target Detection in Maritime Rescue from UAV Imagery |
| title_short | SR-DETR: Target Detection in Maritime Rescue from UAV Imagery |
| title_sort | sr detr target detection in maritime rescue from uav imagery |
| topic | target detection maritime search and rescue drones computer vision |
| url | https://www.mdpi.com/2072-4292/17/12/2026 |
| work_keys_str_mv | AT yulingliu srdetrtargetdetectioninmaritimerescuefromuavimagery AT yanwei srdetrtargetdetectioninmaritimerescuefromuavimagery |