EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Triple-Channel Network for Maritime Radar Targets Detection Based on Multi-Modal Features

Kaiqi Wang, Zeyu Wang

Remote Sensing · 13.12.2024 · 16 (24) · 4662

DOI: 10.3390/rs16244662 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 13.12.2024. Verfasst von Kaiqi Wang und 1 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sea surface target detectors are often interfered by various complex sea surface factors such as sea clutter. Especially when the signal-to-clutter ratio (SCR) is low, it is difficult to achieve high-performance detection. This paper proposes a triple-channel network model for maritime target detection based on the method of multi-modal data fusion. This method comprehensively improves the traditional multi-channel inputs by extracting highly complementary multi-modal features from radar echoes, namely, time-frequency image, phase sequence and correlation coefficient sequence. Appropriate networks are selected to construct a triple-channel network according to the internal data structure of each feature. The three features are utilized as the input of each network channel. To reduce the coupling between multi-channel data, the SE block is introduced to optimize the feature vectors of the channel dimension and improve the data fusion strategy. The detection results are output by the false alarm control unit according to the given probability of false alarm (PFA). The experiments on the IPIX datasets verify that the performance of the proposed detector is better than the existing detectors in dealing with complex ocean scenes.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Wang, K., & Wang, Z. (2024). A Triple-Channel Network for Maritime Radar Targets Detection Based on Multi-Modal Features. Remote Sensing, 16(24), 4662. https://doi.org/10.3390/rs16244662

Vancouver

Wang K, Wang Z. A Triple-Channel Network for Maritime Radar Targets Detection Based on Multi-Modal Features. Remote Sensing. 2024;16(24):4662. doi:10.3390/rs16244662

BibTeX
@article{Wang2024,
  author = {Wang, Kaiqi and Wang, Zeyu},
  title = {A Triple-Channel Network for Maritime Radar Targets Detection Based on Multi-Modal Features},
  journal = {Remote Sensing},
  year = {2024},
  volume = {16},
  number = {24},
  pages = {4662},
  doi = {10.3390/rs16244662}
}
Kostenlos starten →