研究目的
To propose a novel ship segmentation method based on kurtosis test in the complex-valued SAR imagery that can take advantage of the complex information in SAR imagery.
研究成果
The proposed ship segmentation method based on kurtosis test in complex-valued SAR imagery achieves good performance by leveraging the difference in kurtosis between sea clutter and ship targets. The method is robust and can effectively separate sea clutter and ship targets in SAR imagery.
研究不足
The performance of the method may be affected by the quality of SAR imagery and the presence of non-sea clutter objects. The method's robustness across different SAR sensors and imaging conditions needs further validation.
1：Experimental Design and Method Selection:
The method is based on the kurtosis test in complex-valued SAR imagery, leveraging the difference in kurtosis between sea clutter (Gaussian distribution) and ship targets (sup-Gaussian distribution).
2：Sample Selection and Data Sources:
Experimental data are from Gaofen-3 and Sentinel-1 complex-valued data.
3：List of Experimental Equipment and Materials:
Gaofen-3 and Sentinel-1 SAR imagery.
4：Experimental Procedures and Operational Workflow:
The method involves setting an initial threshold, increasing the threshold to compute kurtosis, segmenting using the threshold when kurtosis approaches the ideal value for sea clutter, and outputting binarized images.
5：Data Analysis Methods:
The kurtosis of sea clutter and ship targets is analyzed to validate the segmentation rationale.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
