基于属性散射中心和可分性测度的SAR目标分类方法
By using a method based on attribute scattering centers and separability measures, the problems of insufficient parameter utilization and poor interpretability in existing SAR target classification are solved, achieving more efficient and accurate SAR target identification and enhancing the interpretability of the model.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NAT UNIV OF DEFENSE TECH
- Filing Date
- 2023-10-19
- Publication Date
- 2026-07-17
AI Technical Summary
Existing SAR target classification methods suffer from insufficient parameter utilization, inefficient fusion methods, and poor interpretability, leading to a bottleneck in the development of deep learning applications in SAR target classification.
A method based on attribute scattering centers and separability measures is adopted. By preprocessing complex SAR images, attribute scattering center parameters are extracted and the images are reconstructed. Features are calculated and fused using a separability measure network to guide the model training process, making the features more physically and mathematically interpretable.
It significantly improves the model's recognition efficiency and accuracy, achieves efficient extraction and reliable classification of physical features of SAR images, and enhances the model's physical and mathematical interpretability.
Smart Images
Figure CN117593563B_ABST