Frequency‐Aware Fourier Ptychography for Imaging Beyond Spatial Coherence Constraints

ABSTRACT

Fourier ptychography (FP) is a high‐resolution synthetic‐aperture imaging technique whose standard coherent reconstruction model generally assumes sufficient spatial coherence among aperture‐sampled measurements. In macroscopic imaging, this condition can be weakened by finite source size and scattering‐induced angular broadening, leading to model mismatch, degraded spatial‐frequency selectivity, and reconstruction artifacts. Here, we propose frequency‐aware Fourier ptychography (FA‐FP), a physics‐guided robust reconstruction framework for spatial‐coherence‐degraded illumination. By describing coherence degradation as low‐spatial‐frequency background bias and high‐spatial‐frequency sub‐aperture crosstalk, FA‐FP combines spatial‐domain differential filtering with wavelet‐domain low‐rank regularization to suppress position‐insensitive background components and recover reliable high‐spatial‐frequency structures. Experiments with controlled coherence‐length modulation, variable scanning ranges, transmission and reflection configurations, and diverse samples show that FA‐FP improves reconstruction fidelity and resolution robustness compared with conventional FP under low‐spatial‐coherence conditions. These results establish FA‐FP as a practical robustness‐enhancing strategy for synthetic‐aperture imaging under non‐ideal spatial coherence, opening opportunities for macroscopic FP in long‐distance observation, industrial surface inspection, and reflective microstructure measurement.