Serial coherent diffraction imaging of dynamic samples based on inter-frame constraint

We proposed a novel approach to coherent imaging of dynamic samples. The inter-frame similarity of the sample's local structures is found to be a powerful constraint in phasing a sequence of diffraction patterns. We devised a new image reconstruction algorithm that exploits this inter-frame constraint enabled by an adaptive similar region determination approach. We demonstrated the feasibility of this technique in visible light experiments with various real samples, achieving reconstructions of good quality within a few hundred iterations. With a setup as simple as conventional coherent diffraction imaging but with much-improved convergence and robustness to missing data and noise, our method is expected to enrich X-ray imaging techniques and electron microscopy, offering a new tool for dynamics studies.