Mitigation and quantification of high-frequency instabilities using progressive filtering in burst-mode ptychography

Ptychography is a powerful lensless imaging technique that reconstructs high-resolution images from diffraction patterns. While widely adopted across scientific fields, ptychography is highly sensitive to system instabilities. These instabilities are typically categorized by their frequency relative to the exposure time: low-frequency instabilities, such as beam variations and position errors, can be mitigated effectively, but high-frequency instabilities remain challenging. Here, we present an alternative approach that exploits the burst-mode capability of modern detectors to perform progressive filtering (PF) ptychography, utilizing a temporally oversampled dataset. Applied to a standard Siemens star sample, our method improved the spatial resolution from 30.2 nm to 9.5 nm. Furthermore, we characterize the high-frequency instabilities, revealing a typical length scale of 33 nm and characteristic frequencies of 29 and 38 Hz.