Integrating Prior Knowledge into Attention for Ship Detection in SAR Images
Although they have achieved great success in optical images, deep convolutional neural networks underperform for ship detection in SAR images because of the lack of color and textual features. In this paper, we propose our framework which integrates prior knowledge into neural networks by means of t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/5/2941 |
| _version_ | 1797615777716633600 |
|---|---|
| author | Yin Pan Lei Ye Yingkun Xu Junyi Liang |
| author_facet | Yin Pan Lei Ye Yingkun Xu Junyi Liang |
| author_sort | Yin Pan |
| collection | DOAJ |
| description | Although they have achieved great success in optical images, deep convolutional neural networks underperform for ship detection in SAR images because of the lack of color and textual features. In this paper, we propose our framework which integrates prior knowledge into neural networks by means of the attention mechanism. Because the background of ships is mostly water surface or coast, we use clustering algorithms to generate the prior knowledge map from brightness and density features. The prior knowledge map is later resized and fused with convolutional feature maps by the attention mechanism. Our experiments demonstrate that our framework is able to improve various one-stage and two-stage object detection algorithms (Faster R-CNN, RetinaNet, SSD, and YOLOv4) on two benchmark datasets (SSDD, LS-SSDD, and HRSID). |
| first_indexed | 2024-03-11T07:31:38Z |
| format | Article |
| id | doaj.art-3af32eba27a245539dbbf3f1a3a15163 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T07:31:38Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-3af32eba27a245539dbbf3f1a3a151632023-11-17T07:17:08ZengMDPI AGApplied Sciences2076-34172023-02-01135294110.3390/app13052941Integrating Prior Knowledge into Attention for Ship Detection in SAR ImagesYin Pan0Lei Ye1Yingkun Xu2Junyi Liang3The College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310014, ChinaThe College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310014, ChinaThe College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310014, ChinaThe College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310014, ChinaAlthough they have achieved great success in optical images, deep convolutional neural networks underperform for ship detection in SAR images because of the lack of color and textual features. In this paper, we propose our framework which integrates prior knowledge into neural networks by means of the attention mechanism. Because the background of ships is mostly water surface or coast, we use clustering algorithms to generate the prior knowledge map from brightness and density features. The prior knowledge map is later resized and fused with convolutional feature maps by the attention mechanism. Our experiments demonstrate that our framework is able to improve various one-stage and two-stage object detection algorithms (Faster R-CNN, RetinaNet, SSD, and YOLOv4) on two benchmark datasets (SSDD, LS-SSDD, and HRSID).https://www.mdpi.com/2076-3417/13/5/2941prior knowledgeexpert knowledgeSAR imagesobject detectionattention mechanism |
| spellingShingle | Yin Pan Lei Ye Yingkun Xu Junyi Liang Integrating Prior Knowledge into Attention for Ship Detection in SAR Images Applied Sciences prior knowledge expert knowledge SAR images object detection attention mechanism |
| title | Integrating Prior Knowledge into Attention for Ship Detection in SAR Images |
| title_full | Integrating Prior Knowledge into Attention for Ship Detection in SAR Images |
| title_fullStr | Integrating Prior Knowledge into Attention for Ship Detection in SAR Images |
| title_full_unstemmed | Integrating Prior Knowledge into Attention for Ship Detection in SAR Images |
| title_short | Integrating Prior Knowledge into Attention for Ship Detection in SAR Images |
| title_sort | integrating prior knowledge into attention for ship detection in sar images |
| topic | prior knowledge expert knowledge SAR images object detection attention mechanism |
| url | https://www.mdpi.com/2076-3417/13/5/2941 |
| work_keys_str_mv | AT yinpan integratingpriorknowledgeintoattentionforshipdetectioninsarimages AT leiye integratingpriorknowledgeintoattentionforshipdetectioninsarimages AT yingkunxu integratingpriorknowledgeintoattentionforshipdetectioninsarimages AT junyiliang integratingpriorknowledgeintoattentionforshipdetectioninsarimages |