Speckle Image Restoration without Clean Data
This addresses a practical challenge in fields like digital holography and ultrasound by enabling noise removal without clean data, though it appears incremental as it builds on existing restoration methods.
The paper tackles the problem of speckle noise removal in coherent imaging systems where clean data or multiple noisy observations are unavailable, and demonstrates superior results in both synthetic and real-world digital holography samples compared to baselines.
Speckle noise is an inherent disturbance in coherent imaging systems such as digital holography, synthetic aperture radar, optical coherence tomography, or ultrasound systems. These systems usually produce only single observation per view angle of the same interest object, imposing the difficulty to leverage the statistic among observations. We propose a novel image restoration algorithm that can perform speckle noise removal without clean data and does not require multiple noisy observations in the same view angle. Our proposed method can also be applied to the situation without knowing the noise distribution as prior. We demonstrate our method is especially well-suited for spectral images by first validating on the synthetic dataset, and also applied on real-world digital holography samples. The results are superior in both quantitative measurement and visual inspection compared to several widely applied baselines. Our method even shows promising results across different speckle noise strengths, without the clean data needed.