EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Integrating Target and Shadow Features for SAR Target Recognition

Zhiyuan Zhao, Xiaorong Xue, Iqra Mariam iD, Xing Zhou

Sensors · 22.09.2023 · 23 (19) · 8031

DOI: 10.3390/s23198031 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 22.09.2023. Verfasst von Zhiyuan Zhao und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou 121001, China. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Plan Project of Liaoning Province, China., Key project of scientific research of the Education Department of Liaoning Province, China., Science and Technology Plan Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Synthetic aperture radar (SAR) sensor often produces a shadow in pairs with the target due to its slant-viewing imaging. As a result, shadows in SAR images can provide critical discriminative features for classifiers, such as target contours and relative positions. However, shadows possess unique properties that differ from targets, such as low intensity and sensitivity to depression angles, making it challenging to extract depth features from shadows directly using convolutional neural networks (CNN). In this paper, we propose a new SAR image-classification framework to utilize target and shadow information comprehensively. First, we design a SAR image segmentation method to extract target regions and shadow masks. Second, based on SAR projection geometry, we propose a data-augmentation method to compensate for the geometric distortion of shadows due to differences in depression angles. Finally, we introduce a feature-enhancement module (FEM) based on depthwise separable convolution (DSC) and convolutional block attention module (CBAM), enabling deep networks to fuse target and shadow features adaptively. The experimental results on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset show that when only using target and shadow information, the published deep-learning models can still achieve state-of-the-art performance after embedding the FEM.

Schlagwörter

Stichwörter der Autor:innen

SAR image classificationattention mechanismconvolutional neural network (CNN)features of target and shadowsynthetic aperture radar (SAR)

Autor:innen und Institutionen

  • School of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou 121001, China.

ORCID-Profile: Iqra Mariam

Förderung

  • Science and Technology Plan Project of Liaoning Province, China.
  • Key project of scientific research of the Education Department of Liaoning Province, China.
  • Science and Technology Plan Project
  • Education Department of Liaoning Province, China

Zitieren

APA

Zhao, Z., Xue, X., Mariam, I., & Zhou, X. (2023). Integrating Target and Shadow Features for SAR Target Recognition. Sensors, 23(19), 8031. https://doi.org/10.3390/s23198031

Vancouver

Zhao Z, Xue X, Mariam I, Zhou X. Integrating Target and Shadow Features for SAR Target Recognition. Sensors. 2023;23(19):8031. doi:10.3390/s23198031

BibTeX
@article{Zhao2023,
  author = {Zhao, Zhiyuan and Xue, Xiaorong and Mariam, Iqra and Zhou, Xing},
  title = {Integrating Target and Shadow Features for SAR Target Recognition},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {19},
  pages = {8031},
  doi = {10.3390/s23198031}
}
Kostenlos starten →