A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR
Compact High frequency surface wave radar (HFSWR) has been widely used in remote sensing of oceanic dynamics and ship targets due to its convenient deployment and low cost. However, when using a constant false alarm rate (CFAR) detector, these systems experience performance degradation primarily bec...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10804208/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841557089628979200 |
|---|---|
| author | Da Huang Hao Zhou Yingwei Tian Zhiqing Yang Weimin Huang |
| author_facet | Da Huang Hao Zhou Yingwei Tian Zhiqing Yang Weimin Huang |
| author_sort | Da Huang |
| collection | DOAJ |
| description | Compact High frequency surface wave radar (HFSWR) has been widely used in remote sensing of oceanic dynamics and ship targets due to its convenient deployment and low cost. However, when using a constant false alarm rate (CFAR) detector, these systems experience performance degradation primarily because of echo nonstationarity. To address this challenge, a deep learning (DL)-based scheme tailored for identifying ship targets in the time-frequency (TF) domain is presented. To ensure high-quality model training, we develop a semiautomatic annotation approach that uses automatic identification system (AIS) information as a reference and collect a TF dataset named HFSWR-TFD. In addition, inspired by the dynamic snake convolution and triplet attention mechanism, an improved YOLOv5s model named DS-YOLOv5s is designed to effectively capture target ridges. The inference results are filtered using a confidence threshold and then transformed into the range-Doppler domain for final target identification. Experimental results on the newly collected dataset show significant improvements are achieved by DS-YOLOv5s. Compared to its baseline, the DS-YOLOv5s can increase the F1 score by 15.3%, and AP75 by 6.3%. Then, this pretrained DL model is integrated into the entire scheme to make comparison with existing CFAR detectors. With the AIS records as ground truth, our scheme achieves a match rate that is 2.27<inline-formula><tex-math notation="LaTeX">$\sim$</tex-math></inline-formula>8.17% greater than its CFAR counterparts. Moreover, the quantitative results of the associate tracks further confirm the superiority of the proposed method. In conclusion, the proposed scheme provides an effective and efficient solution for HFSWR ship detection. |
| format | Article |
| id | doaj-art-7ada3b09d2144fb2b7b0d935a6cdc4c3 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-7ada3b09d2144fb2b7b0d935a6cdc4c32025-01-07T00:00:30ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01182718273610.1109/JSTARS.2024.351878110804208A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWRDa Huang0https://orcid.org/0000-0002-0373-9180Hao Zhou1https://orcid.org/0000-0003-3680-5903Yingwei Tian2https://orcid.org/0000-0003-3450-3688Zhiqing Yang3https://orcid.org/0009-0008-3579-8901Weimin Huang4https://orcid.org/0000-0001-9622-5041School of Electronic Information, Wuhan University, Wuhan, ChinaSchool of Electronic Information, Wuhan University, Wuhan, ChinaSchool of Electronic Information, Wuhan University, Wuhan, ChinaCollege of Information Science and Engineering, Henan University of Technology, Zhengzhou, ChinaFaculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John's, NL, CanadaCompact High frequency surface wave radar (HFSWR) has been widely used in remote sensing of oceanic dynamics and ship targets due to its convenient deployment and low cost. However, when using a constant false alarm rate (CFAR) detector, these systems experience performance degradation primarily because of echo nonstationarity. To address this challenge, a deep learning (DL)-based scheme tailored for identifying ship targets in the time-frequency (TF) domain is presented. To ensure high-quality model training, we develop a semiautomatic annotation approach that uses automatic identification system (AIS) information as a reference and collect a TF dataset named HFSWR-TFD. In addition, inspired by the dynamic snake convolution and triplet attention mechanism, an improved YOLOv5s model named DS-YOLOv5s is designed to effectively capture target ridges. The inference results are filtered using a confidence threshold and then transformed into the range-Doppler domain for final target identification. Experimental results on the newly collected dataset show significant improvements are achieved by DS-YOLOv5s. Compared to its baseline, the DS-YOLOv5s can increase the F1 score by 15.3%, and AP75 by 6.3%. Then, this pretrained DL model is integrated into the entire scheme to make comparison with existing CFAR detectors. With the AIS records as ground truth, our scheme achieves a match rate that is 2.27<inline-formula><tex-math notation="LaTeX">$\sim$</tex-math></inline-formula>8.17% greater than its CFAR counterparts. Moreover, the quantitative results of the associate tracks further confirm the superiority of the proposed method. In conclusion, the proposed scheme provides an effective and efficient solution for HFSWR ship detection.https://ieeexplore.ieee.org/document/10804208/Deep learning (DL)dynamic snake convolution (DSConv)high-frequency surface wave radar (HFSWR)ship detectiontime-frequency analysis (TFA) |
| spellingShingle | Da Huang Hao Zhou Yingwei Tian Zhiqing Yang Weimin Huang A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning (DL) dynamic snake convolution (DSConv) high-frequency surface wave radar (HFSWR) ship detection time-frequency analysis (TFA) |
| title | A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR |
| title_full | A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR |
| title_fullStr | A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR |
| title_full_unstemmed | A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR |
| title_short | A Deep Learning-Based Time-Frequency Scheme for Ship Detection Using HFSWR |
| title_sort | deep learning based time frequency scheme for ship detection using hfswr |
| topic | Deep learning (DL) dynamic snake convolution (DSConv) high-frequency surface wave radar (HFSWR) ship detection time-frequency analysis (TFA) |
| url | https://ieeexplore.ieee.org/document/10804208/ |
| work_keys_str_mv | AT dahuang adeeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT haozhou adeeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT yingweitian adeeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT zhiqingyang adeeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT weiminhuang adeeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT dahuang deeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT haozhou deeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT yingweitian deeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT zhiqingyang deeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr AT weiminhuang deeplearningbasedtimefrequencyschemeforshipdetectionusinghfswr |