Abstract:Coherent diffraction imaging (CDI) and its scanning version, ptychography, are lensless imaging approaches used to iteratively retrieve a sample's complex scattering amplitude from its measured diffraction patterns. These imaging methods are most useful in extreme ultraviolet (EUV) and X-ray regions of the electromagnetic spectrum, where efficient imaging optics are difficult to manufacture. CDI relies on high signal-to-noise ratio diffraction data to recover the phase, but increasing the flux can cause saturation effects on the detector. A conventional solution to this problem is to place a beam stop in front of the detector. The pixel masking method is a common solution to the problem of missing frequencies due to a beam stop. This paper describes the information redundancy in the recorded data set and expands on how the reconstruction algorithm can exploit this redundancy to estimate the missing frequencies. Thereafter, we modify the size of the beam stop in experimental and simulation data to assess the impact of the missing frequencies, investigate the extent to which the lost portion of the diffraction spectrum can be recovered, and quantify the effect of the beam stop on the image quality. The experimental findings and simulations conducted for EUV imaging demonstrate that when using a beam stop, the numerical aperture of the condenser is a crucial factor in the recovery of lost frequencies. Our thorough investigation of the reconstructed images provides information on the overall quality of reconstruction and highlights the vulnerable frequencies if the beam stop size is larger than the extent of the illumination NA. The outcome of this study can be applied to other sources of frequency loss, and it will contribute to the improvement of experiments and reconstruction algorithms in CDI.

Centroid adapted frequency selective extrapolation for reconstruction of lost image areas

Modified Edge-Oriented Spatial Interpolation For Consecutive Blocks Error Concealment

Projection Data Recovery Induced Non-local Means for Low-Dose CT Reconstruction

Edge Guided Reconstruction for Compressive Imaging

Spatio-spectral Image Reconstruction Using Non-local Filtering

Motion Compensated Three-Dimensional Frequency Selective Extrapolation for Improved Error Concealment in Video Communication

Resampling Images to a Regular Grid from a Non-Regular Subset of Pixel Positions Using Frequency Selective Reconstruction

Spatial error concealment based on line segmentation and adaptive edge points matching

Sequential Image Recovery from Noisy and Under-Sampled Fourier Data

Selective pixel interpolation for spatial error concealment

Bandwidth-insensitive Extended Centroid Frequency Shift Method for Near-Surface Q Estimation

Extrapolative Phase Retrieval Based on a Hybrid of PhaseCut and Alternating Projection Techniques

Edge Re-Projection Method for High-Quality Edge Reconstruction in Non-Line-of-sight Imaging.

Image Reconstruction in Peripheral and Central Regions-of-interest and Data Redundancy.

Recovery of spatial frequencies in coherent diffraction imaging in the presence of a central obscuration

A New Extrapolation Economy Cascadic Multigrid Method for Image Restoration Problems

Globally convergent edge-preserving reconstruction with contour-line smoothing

ADAPTIVEPIXELINTERPOLATIONFOR SPATIALERROR CONCEALMENT

Image Reconstruction of Sparse Fan-Beam Projections Using a Hybrid Algorithm

Spectrum-estimation Based Lossless Information Recovery for Sparse Array Patterns

An Accurate Centroiding Algorithm for PSF Reconstruction