A Novel Fusion-Based Ship Detection Method from Pol-SAR Images

Area of Science:

• Remote Sensing
• Radar Imaging
• Signal Processing

Background:

• Ship detection in polarimetric Synthetic Aperture Radar (Pol-SAR) images is crucial for maritime surveillance.
• Existing methods may struggle with accuracy and preserving ship shape.
• Integrating multiple features can potentially improve detection performance.

Purpose of the Study:

• To propose a novel fusion-based ship detection method for Pol-SAR images.
• To enhance the accuracy of ship detection and improve ship shape preservation.
• To validate the proposed method against established techniques.

Main Methods:

• Feature extraction and Constant False Alarm Rate (CFAR) detection were performed.
• Detection results from HH channel, Pauli decomposition (diplane scattering), and Barnes decomposition (helical factor) were fused.
• Potential target pixels were classified using target difference degree.

Main Results:

• The fusion-based method successfully identified confirmed and potential ship targets.
• The method demonstrated superior performance in both ship detection and shape preservation.
• Comparative analysis showed better results than K-means clustering and Notch Filter methods.

Conclusions:

• The proposed fusion-based ship detection method effectively utilizes multiple features for accurate results.
• This novel approach offers improved performance for ship detection and shape preservation in Pol-SAR imagery.
• The method shows significant potential for maritime surveillance applications.