电子学报2025，Vol.53Issue(9)：3147-3162,16.DOI:10.12263/DZXB.20250417
跨模态渐进式知识迁移SAR目标检测技术
Cross-Modal SAR Target Detection via Progressive Knowledge Transfer
摘要
Abstract
The wide application of electro-optical sensors presented the urgent need of cross-modal learning between optical and SAR image.In this paper,a new cross-modal synthetic aperture radar(SAR)target detection method via progres-sive knowledge transfer was proposed.First,a new generative technique from the optical image to SAR image was present-ed.The immediate domain composed of the generated pseudo SAR images can be formed accordingly.The semantic dis-crepancies between SAR backscattering imaging mechanism and the passive optical radiation imagery can be bridged.The optical radiation features with SAR scattering characteristics can be fused effectively.Second,a dual-stage domain adapta-tion strategy composed of the multi-scale feature alignment skill was presented.The semantic components between the opti-cal source domain and the intermediate domain can be aligned through the multi-scale feature learning trick initially.The scattering distribution alignment between the intermediate domain and the SAR target domain can be then achieved.Third,a quality-aware dynamic weighting strategy was presented to mitigate the impact of outlier samples in the intermediate do-main.It was capable of adjusting the contributions of synthetic data based on confidence metrics dynamically.Finally,mul-tiple rounds of experiments were pursued on SpaceNet6,SSDD(SAR Ship Detection Dataset),and HRSID(High-Resolu-tion SAR Images Dataset)datasets.The results proved the advantages of proposed method.The improvement of 21.5 per-centage points was achieved compared to the source-only learning method.Likewise,the improvement of 3.3 percentage points was achieved in comparison to the state-of-the-art.These results confirm the viability of electro-optical-to-SAR knowledge transfer for enhancing cross-modal target detection.关键词
合成孔径雷达/目标检测/知识迁移/对抗学习/多尺度对齐/伪标签学习Key words
synthetic aperture radar/target detection/knowledge transfer/adversarial learning/multi-scale align-ment/pseudo-label learning分类
信息技术与安全科学引用本文复制引用
赵国威,蒋嘉庆,董刚刚..跨模态渐进式知识迁移SAR目标检测技术[J].电子学报,2025,53(9):3147-3162,16.基金项目
国家自然科学基金(No.61971324,No.62525105) National Natural Science Foundation of China(No.61971324,No.62525105) （No.61971324,No.62525105）
