Domain Adaptive SAR Wake Detection: Leveraging Similarity Filtering and Memory Guidance

2025-09-14

AI Summary

Optical-image-trained wake detection models suffer from severe domain shift when directly applied to noisy, abstract SAR imagery. To address this challenge, we propose SimMemDA, a cross-modal unsupervised domain adaptation framework for optical-to-SAR wake detection. Methodologically, it integrates WakeGAN-based style transfer to reduce modality discrepancy; introduces an instance-level feature similarity filtering mechanism to mitigate negative transfer; and constructs a dynamically updated feature–confidence memory bank, which refines pseudo-labels via K-nearest-neighbor weighting and region-mixing calibration. Experiments demonstrate that SimMemDA significantly improves detection accuracy and robustness on SAR images, consistently outperforming state-of-the-art unsupervised domain adaptation methods across multiple SAR datasets. This validates the effectiveness and feasibility of our cross-modal adaptive approach for wake detection.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Web data generation and simulation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Synthetic Aperture Radar (SAR), with its all- weather and wide-area observation capabilities, serves as a crucial tool for wake detection. However, due to its complex imaging mechanism, wake features in SAR images often appear abstract and noisy, posing challenges for accurate annotation. In contrast, optical images provide more distinct visual cues, but models trained on optical data suffer from performance degradation when applied to SAR images due to domain shift. To address this cross-modal domain adaptation challenge, we propose a Similarity-Guided and Memory-Guided Domain Adap- tation (termed SimMemDA) framework for unsupervised domain adaptive ship wake detection via instance-level feature similarity filtering and feature memory guidance. Specifically, to alleviate the visual discrepancy between optical and SAR images, we first utilize WakeGAN to perform style transfer on optical images, generating pseudo-images close to the SAR style. Then, instance-level feature similarity filtering mechanism is designed to identify and prioritize source samples with target-like dis- tributions, minimizing negative transfer. Meanwhile, a Feature- Confidence Memory Bank combined with a K-nearest neighbor confidence-weighted fusion strategy is introduced to dynamically calibrate pseudo-labels in the target domain, improving the reliability and stability of pseudo-labels. Finally, the framework further enhances generalization through region-mixed training, strategically combining source annotations with calibrated tar- get pseudo-labels. Experimental results demonstrate that the proposed SimMemDA method can improve the accuracy and robustness of cross-modal ship wake detection tasks, validating the effectiveness and feasibility of the proposed method.

Problem

• Cross-modal domain adaptation for SAR wake detection
• Minimizing negative transfer between optical and SAR images
• Improving pseudo-label reliability in unsupervised domain adaptation

Innovation

• Style transfer with WakeGAN for SAR-like images
• Instance-level similarity filtering to minimize negative transfer
• Feature-confidence memory bank for dynamic pseudo-label calibration