• Journal of Infrared and Millimeter Waves
  • Vol. 41, Issue 3, 618 (2022)
Li-Yuan LI1、3, Xiao-Yan LI2, Zhuo-Yue HU1, Xiao-Feng SU1、*, and Fan-Sheng CHEN1、2、*
Author Affiliations
  • 1Key Laboratory of Intelligent Infrared Perception,Shanghai Institute of Technical Physics,Chinese Academy of Sciences,Shanghai 200083,China
  • 2Hangzhou Institute for Advanced Study,University of Chinese Academy of Sciences,Hangzhou 310024,China
  • 3University of Chinese Academy of Sciences,Beijing 100049,China
  • show less
    DOI: 10.11972/j.issn.1001-9014.2022.03.013 Cite this Article
    Li-Yuan LI, Xiao-Yan LI, Zhuo-Yue HU, Xiao-Feng SU, Fan-Sheng CHEN. The research on lightweight SAR ship detection method based on regression model and attention[J]. Journal of Infrared and Millimeter Waves, 2022, 41(3): 618 Copy Citation Text show less

    Abstract

    Synthetic aperture radar (SAR) has the advantages of all-sky and all-weather earth observation without cloud interference. Ship detection based on SAR images has been widely used in civil and military fields, including maritime search and rescue, port reconnaissance, territorial sea defense. However, different from large ships, the misdetection rate of small ships with fewer pixels and lower contrast is high. And it is difficult to balance speed and accuracy during on-orbit ship detection. To solve the above problems, an improved lightweight ships detection method (ImShips) based on YOLOv5s is proposed. Firstly, the standard convolution with small receptive field is adopted at the bottom of the baseline to obtain spatial information about small ships. And the dilated convolution with enlarged receptive field is added at the top of the baseline to preserve more semantic features, which is conducive to extract large targets feature. Then, a lightweight channel attention mechanism is applied to the backbone and neck of YOLOv5. And the weight is allocated to filter more important texture information. Finally, the depth-wise separable convolution is adopted to replace the standard convolution during down-sampling to reduce the number of parameters and improve the inference speed. Compared with YOLOv5s model, the experimental results show that ImShips achieve an increase in AP, while the FLOPs are reduced by 45.61%, and the speed is increased by 8.31% in SSDD and ISSID datasets. The speed and accuracy of ImShips model are improved effectively on sea surface object detection. The proposed method has great application potential for on-orbit ship detection.