Holographic imaging poses significant challenges when facing real-time disturbances introduced by dynamic environments. The existing deep-learning methods for holographic imaging often depend solely on the specific condition based on the given data distributions, thus hindering their generalization across multiple scenes. One critical problem is how to guarantee the alignment between any given downstream tasks and pretrained models. We analyze the physical mechanism of image degradation caused by turbulence and innovatively propose a swin transformer-based method, termed train-with-coherence-swin (TWC-Swin) transformer, which uses spatial coherence (SC) as an adaptable physical prior information to precisely align image restoration tasks in the arbitrary turbulent scene. The light-processing system (LPR) we designed enables manipulation of SC and simulation of any turbulence. Qualitative and quantitative evaluations demonstrate that the TWC-Swin method presents superiority over traditional convolution frameworks and realizes image restoration under various turbulences, which suggests its robustness, powerful generalization capabilities, and adaptability to unknown environments. Our research reveals the significance of physical prior information in the optical intersection and provides an effective solution for model-to-tasks alignment schemes, which will help to unlock the full potential of deep learning for all-weather optical imaging across terrestrial, marine, and aerial domains.

将相变介质的有源调控属性与基于广义斯涅尔定律的相位型超表面相结合，实现了可重构的近红外全息超构器件。当集成的锗锑碲化合物（GST）相变介质处于非晶态时，各向异性的超构单元可实现宽带范围内（1.55~2.8 μm）交叉极化转化率为~80%的几何相位调控，进而实现近场全息相位生成以及远场全息成像，即器件处于“开”态。当GST相变为晶态时，超构单元交叉极化效率被极小化，全息相位以及远场全息图像被关闭，器件处于“关”态。本文提出的相位型开关器件将在空间光调制、波前工程、全息成像等有源光子及光电集成应用中展现出极大潜力。

Fresnel incoherent correlation holography (FINCH) is a well-established incoherent imaging technique. In FINCH, three self-interference holograms are recorded with calculated phase differences between the two interfering, differently modulated object waves and projected into a complex hologram. The object is reconstructed without the twin image and bias terms by a numerical Fresnel back propagation of the complex hologram. A modified approach to implement FINCH by a single camera shot by pre-calibrating the system involving recording of the point spread function library and reconstruction by a non-linear cross correlation has been introduced recently. The expression of the imaging characteristics from the modulation functions in original FINCH and the modified approach by pre-calibration in spatial and polarization multiplexing schemes are reviewed. The study reveals that a reconstructing function completely independent of the function of the phase mask is required for the faithful expression of the characteristics of the modulating function in image reconstruction. In the polarization multiplexing method by non-linear cross correlation, a partial expression was observed, while in the spatial multiplexing method by non-linear cross correlation, the imaging characteristics converged towards a uniform behavior.