Influence of Noise and Missing Data on Reconstruction Quality in Coherent X-ray Diffractive Imaging
Abstract
Coherent x-ray diffractive imaging (CXDI) is a microscopy technique that numerically inverts measured diffraction intensities to arrive at the complex index of refraction of the specimen under investigation. We explore here the performance of scanning CXDI using the Difference Map algorithm when using diffraction with degraded signal-to-noise ratio (SNR) and diffraction with missing regions, due to, for example, a beam stop. We quantify reconstruction quality as a function of SNR using the concept of the phase retrieval transfer function which seeks to identify a cutoff spatial resolution up to which we can trust spatial features in the image. We also explore reconstruction degradation as a function of missing data region size.
- Publication:
-
10th International Conference on X-ray Microscopy
- Pub Date:
- September 2011
- DOI:
- 10.1063/1.3625365
- Bibcode:
- 2011AIPC.1365..305T
- Keywords:
-
- coherence;
- X-ray microscopy;
- image reconstruction;
- numerical analysis;
- 42.25.Kb;
- 68.37.Yz;
- 42.30.Wb;
- 02.60.Cb;
- Coherence;
- X-ray microscopy;
- Image reconstruction;
- tomography;
- Numerical simulation;
- solution of equations