Correlation adaptive optics enables label-free imaging through distortion

Optical instruments that observe through a medium, such as telescopes or microscopes, can have their images distorted by the viewing medium or the optical components themselves. To prevent this, adaptive optics systems continuously adjust their optical elements, compensating for distortion in real time. Typically, this is done by tracking a “guide star,” a single point of light — literally a star, in the case of telescopes — and using its apparent fluctuations to make adjustments.

However, in some cases, such as label-free microscopy, guide stars are unavailable, requiring alternative approaches. He et al. developed correlation adaptive optics (CAO) for label-free wavefront correction as a solution to this problem.

Their approach relies on generating even-symmetrical thermal light (ETL), which is produced by impinging a coherent beam of light onto a spatial light modulator. When this light is passed through a distorting medium, the symmetry breaks in a measurable way.

“By generating ETL and employing the sum-projection of intensity correlations as a feedback metric, the method iteratively corrects optical aberrations,” said author Jun Xiong.

The authors demonstrated its effectiveness with experiments, resolving an image under heavy distortion and object occlusion — where the target is partially obscured by another object.

Because this technique can be applied as a classical or quantum approach, it can function even in conditions with an extremely low photon flux.

“This work could advance the development of computational imaging systems, such as structured illumination microscopy or single-pixel imaging, by providing a simple, non-invasive strategy for aberration correction,” said Xiong.

The authors plan to expand their approach to include other types of symmetry and incorporate their technique into practical computational imaging systems.

Source: “Label-free correlation adaptive optics using even-symmetrical thermal light,” by Run-Jie He, Zhiyuan Ye, Hongchao Liu, Hai-Bo Wang, and Jun Xiong, APL Photonics (2025). The article can be accessed at https://doi.org/10.1063/5.0283659 .