CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection
Infrared ocean ships detection still faces great challenges due to the low signal-to-noise ratio and low spatial resolution resulting in a severe lack of texture details for small infrared targets, as well as the distribution of the extremely multiscale ships. In this paper, we propose a CAA-YOLO to...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/10/3782 |
| _version_ | 1827666584593760256 |
|---|---|
| author | Jing Ye Zhaoyu Yuan Cheng Qian Xiaoqiong Li |
| author_facet | Jing Ye Zhaoyu Yuan Cheng Qian Xiaoqiong Li |
| author_sort | Jing Ye |
| collection | DOAJ |
| description | Infrared ocean ships detection still faces great challenges due to the low signal-to-noise ratio and low spatial resolution resulting in a severe lack of texture details for small infrared targets, as well as the distribution of the extremely multiscale ships. In this paper, we propose a CAA-YOLO to alleviate the problems. In this study, to highlight and preserve features of small targets, we apply a high-resolution feature layer (P2) to better use shallow details and the location information. In order to suppress the shallow noise of the P2 layer and further enhance the feature extraction capability, we introduce a TA module into the backbone. Moreover, we design a new feature fusion method to capture the long-range contextual information of small targets and propose a combined attention mechanism to enhance the ability of the feature fusion while suppressing the noise interference caused by the shallow feature layers. We conduct a detailed study of the algorithm based on a marine infrared dataset to verify the effectiveness of our algorithm, in which the AP and AR of small targets increase by 5.63% and 9.01%, respectively, and the mAP increases by 3.4% compared to that of YOLOv5. |
| first_indexed | 2024-03-10T01:52:43Z |
| format | Article |
| id | doaj.art-31d43dac39bb4f808c7d060e66a4123c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T01:52:43Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-31d43dac39bb4f808c7d060e66a4123c2023-11-23T13:01:15ZengMDPI AGSensors1424-82202022-05-012210378210.3390/s22103782CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships DetectionJing Ye0Zhaoyu Yuan1Cheng Qian2Xiaoqiong Li3School of Life Science, Beijing Institute of Technology, Beijing 100081, ChinaKey Laboratory of Convergence Medical Engineering System and Healthcare Technology, Ministry of Industry and Information Technology, Beijing 100081, ChinaSchool of Life Science, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Life Science, Beijing Institute of Technology, Beijing 100081, ChinaInfrared ocean ships detection still faces great challenges due to the low signal-to-noise ratio and low spatial resolution resulting in a severe lack of texture details for small infrared targets, as well as the distribution of the extremely multiscale ships. In this paper, we propose a CAA-YOLO to alleviate the problems. In this study, to highlight and preserve features of small targets, we apply a high-resolution feature layer (P2) to better use shallow details and the location information. In order to suppress the shallow noise of the P2 layer and further enhance the feature extraction capability, we introduce a TA module into the backbone. Moreover, we design a new feature fusion method to capture the long-range contextual information of small targets and propose a combined attention mechanism to enhance the ability of the feature fusion while suppressing the noise interference caused by the shallow feature layers. We conduct a detailed study of the algorithm based on a marine infrared dataset to verify the effectiveness of our algorithm, in which the AP and AR of small targets increase by 5.63% and 9.01%, respectively, and the mAP increases by 3.4% compared to that of YOLOv5.https://www.mdpi.com/1424-8220/22/10/3782small targets detectioncombined attention mechanismmultiscale feature fusioninfrared imagemultiscale objects |
| spellingShingle | Jing Ye Zhaoyu Yuan Cheng Qian Xiaoqiong Li CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection Sensors small targets detection combined attention mechanism multiscale feature fusion infrared image multiscale objects |
| title | CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection |
| title_full | CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection |
| title_fullStr | CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection |
| title_full_unstemmed | CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection |
| title_short | CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection |
| title_sort | caa yolo combined attention augmented yolo for infrared ocean ships detection |
| topic | small targets detection combined attention mechanism multiscale feature fusion infrared image multiscale objects |
| url | https://www.mdpi.com/1424-8220/22/10/3782 |
| work_keys_str_mv | AT jingye caayolocombinedattentionaugmentedyoloforinfraredoceanshipsdetection AT zhaoyuyuan caayolocombinedattentionaugmentedyoloforinfraredoceanshipsdetection AT chengqian caayolocombinedattentionaugmentedyoloforinfraredoceanshipsdetection AT xiaoqiongli caayolocombinedattentionaugmentedyoloforinfraredoceanshipsdetection |