Denoising Poisson Phaseless Measurements via Orthogonal Dictionary Learning
arXiv:1612.08656 · doi:10.1364/OE.26.019773
Abstract
Phaseless diffraction measurements recorded by a CCD detector are often affected by Poisson noise. In this paper, we propose a dictionary learning model by employing patches based sparsity to denoise Poisson phaseless measurement. The model consists of three terms: (i) A representation term by an orthogonal dictionary, (ii) an pseudo norm of coefficient matrix, and (iii) a Kullback-Leibler divergence to fit phaseless Poisson data. Fast Alternating Minimization Method (AMM) and Proximal Alternating Linearized Minimization (PALM) are adopted to solve the established model with convergence guarantee, and especially global convergence for PALM is derived. The subproblems for two algorithms have fast solvers, and indeed, the solutions for the sparse coding and dictionary updating both have closed forms due to the orthogonality of learned dictionaries. Numerical experiments for phase retrieval using coded diffraction and ptychographic patterns are performed to show the efficiency and robustness of proposed methods, which, by preserving texture features, produce visually and quantitatively improved denoised images compared with other phase retrieval algorithms without regularization and local sparsity promoting algorithms.
arXiv comment14 pages, 11 figures
References in corpus (17)
- Femtosecond Diffractive Imaging with a Soft-X-ray Free-Electron Laser
- X-ray image reconstruction from a diffraction pattern alone
- A unified evaluation of iterative projection algorithms for phase retrieval
- Relaxed Averaged Alternating Reflections for Diffraction Imaging
- GESPAR: Efficient Phase Retrieval of Sparse Signals
- Compressive Phase Retrieval via Generalized Approximate Message Passing
- Alternating Projection, Ptychographic Imaging and Phase Synchronization
- Sparsity Based Poisson Denoising with Dictionary Learning
- DOLPHIn - Dictionary Learning for Phase Retrieval
- Phase Retrieval Meets Statistical Learning Theory: A Flexible Convex Relaxation
- Undersampled Phase Retrieval via Majorization-Minimization
- Sparse Signal Recovery from Quadratic Measurements via Convex Programming
- Robust Compressive Phase Retrieval via L1 Minimization With Application to Image Reconstruction
- Sparse Phase Retrieval via Truncated Amplitude Flow
- Fourier Phase Retrieval with a Single Mask by Douglas-Rachford Algorithm
- Robust Sparse Phase Retrieval Made Easy
- -regularized Variational Methods for Sparse Phase Retrieval