Physics-Guided Self-Supervised Statistical Residual Learning for Sonar Despeckling with Improved Generalization
For sonar imaging applications, this work provides a practical despeckling solution that eliminates the need for clean reference images while maintaining high performance and generalization.
The paper introduces a self-supervised sonar despeckling method that uses physics-guided constraints in the log domain to suppress speckle without clean training data, achieving state-of-the-art results across multiple real sonar datasets with cross-dataset robustness and real-time suitability.
This letter introduces a physics-informed self-supervised framework for sonar image despeckling that reformulates despeckling as residual consistency in the homomorphic log domain. By constraining the log-ratio residual to obey multiplicative speckle statistics, the proposed method eliminates the need for clean supervision while preventing degenerate identity solutions. A variance-targeted statistical loss combined with edge-aware structural regularization and median-guided curriculum stabilization enables effective speckle suppression with preserved structural fidelity. This formulation along with a lightweight neural network achieves state-of-the-art performance across multiple real sonar datasets and demonstrates excellent cross-dataset robustness, while remaining suitable for real-time deployment.