Adaptive optics (AO) systems compensate for image distortions that can occur when an object is viewed through a microscope, telescope, or other optical instrument. An AO system will adjust the instrument’s optical elements continuously, correcting aberrations in real time. Typically, an AO system detects distortions by tracking a guide star — a single point of light that, for telescopes, is an actual star. The AO system analyzes apparent fluctuations in the light of the guide star and adjusts the optical imager to offset these shifts. The AO system works iteratively, allowing the optical instrument to continuously adapt to changing conditions. However, some optical instruments, such as label-free microscopes, do not have access to a guide star. To address the need for AO without a guide star, researchers at Beijing Normal University and the University of Macau developed correlation adaptive optics (CAO), a guide-star-free wavefront correction technique. Correlation adaptive optics (CAO) enables label-free adaptive optical imaging. Courtesy of Beijing Normal University. Quantum-inspired CAO relies on symmetry breaking to allow wavefront correction without the need for a guide star. It leverages intensity correlations in even-symmetrical thermal light (ETL). The researchers produce even-symmetrical thermal light by using a coherent beam to impinge onto a spatial light modulator (SLM). When the light is passed through a distorting medium, the symmetry breaks in a measurable way. The correlation strength of the intensity correlation function of the distorted light behaves as a feedback metric. “By generating [even-symmetrical thermal light] and employing the sum-projection of intensity correlations as a feedback metric, the method iteratively corrects optical aberrations,” professor Jun Xiong said. By using the anti-correlation property in the symmetric random light, the researchers can directly optimize the distorted point spread function of the imaging system without the need for guide stars or a complex optimization algorithm. In experiments, the team used CAO to resolve an image under heavy distortion and object occlusion. CAO was effective at correcting aberrations, even when the image target was partially obscured by another object. The researchers successfully removed common low-order aberrations, including tilt, astigmatism, defocus, and misalignment, using CAO. CAO was found to perform well on both artificial and natural distortions. The CAO technique can be used as a classical or quantum optical approach. It functions effectively even in conditions where there is an extremely low photon flux. Compared to entangled photon pairs produced in spontaneous parametric down-conversion, the even-symmetrical thermal light used in the CAO scheme is much easier to generate with high brightness using a commercial SLM. This advantage could lead to the use of CAO for adaptive aberration correction in computational imaging systems that use structured illumination. The team believes that its label-free correction process for optical aberrations could be extended to various computational imaging modalities that are based on SLMs and digital micromirror devices, to eliminate potential aberrations in these systems. The optical components of these systems overlap considerably with the CAO scheme, differing mainly in their auxiliary lens groups. “This work could advance the development of computational imaging systems, such as structured illumination microscopy or single-pixel imaging, by providing a simple, noninvasive strategy for aberration correction,” Xiong said. The researchers plan to expand their approach to AO to include other types of symmetry. They also intend to incorporate the CAO technique into practical computational imaging systems. The research was published in APL Photonics (www.doi.org/10.1063/5.0283659).