Fig. 4. Impact of the intensity level on the inversion quality (numerical study).

The 3D Bragg ptychography data set was simulated with 100 times less intensity (maximum of three photons per pixel) with respect to the experimental data and poised with Poisson noise. The sample is strain-free (constant phase) and the scanning parameters are the same as the ones of the experimental data set. In a–d no spatial drift was considered (perfect setup) and a step-scan mode was introduced to generate the data. In e–h, spatial drifts and fly-scan were introduced to generate the data and the new inversion approach was used to retrieve the object. The comparison with a–d shows a strong degradation of the image, due to the impossibility to retrieve in a satisfactory manner the object, the probe, and the sample position, as a result of a lack of information in the data set (limited intensity level)