Two-layer Residual Sparsifying Transform Learning for Image Reconstruction
This work addresses image reconstruction challenges in medical imaging, but it appears incremental as it extends existing transform models with an additional layer.
The paper tackles the problem of image reconstruction from limited and corrupted data, such as in low-dose CT, by proposing a two-layer sparsifying transform model that further sparsifies residuals, showing preliminary numerical usefulness over previous schemes.
Signal models based on sparsity, low-rank and other properties have been exploited for image reconstruction from limited and corrupted data in medical imaging and other computational imaging applications. In particular, sparsifying transform models have shown promise in various applications, and offer numerous advantages such as efficiencies in sparse coding and learning. This work investigates pre-learning a two-layer extension of the transform model for image reconstruction, wherein the transform domain or filtering residuals of the image are further sparsified in the second layer. The proposed block coordinate descent optimization algorithms involve highly efficient updates. Preliminary numerical experiments demonstrate the usefulness of a two-layer model over the previous related schemes for CT image reconstruction from low-dose measurements.