Deep learning wavefront sensing method for Shack-Hartmann sensors with sparse sub-apertures

He, Yunbo; Liu, Zhiwei; Ning, Yu; Li, Jun; Xu, Xiaojun; Jiang, Zongfu

doi:10.1364/oe.427261

Cited by 26 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reconstruction may be further accelerated via parallel computing or via more advanced algorithms. [ 31 ]…”

Section: Resultsmentioning

confidence: 99%

Single‐Shot Wavefront Sensing with Nonlocal Thin Film Optical Filters

Li,

Jia,

Jin

et al. 2023

Laser & Photonics Reviews

View full text Add to dashboard Cite

Conventional wavefront sensors suffer from the fundamental limitation of the space‐bandwidth product and have a trade‐off between their spatial sampling interval and dynamic range. Here, nonlocal thin film optical filters with optimized angle‐ and polarization‐dependent responses are leveraged to circumvent the fundamental limitation, and a robust single‐shot wavefront sensing system with a small spatial sampling interval of 6.9 µm and a large angular dynamic range of 15° is realized. The system only requires inserting two multilayer dielectric filters, fabricated using a mature thin film deposition technique, into a conventional 4‐f imaging apparatus. The polarization‐sensitive nonlocal filters are used to map the 2D phase gradients of the incident light field to the intensity variation of the x‐ and y‐polarized light, respectively, thus enabling single‐shot 2D wavefront reconstruction from images taken by a polarization camera. Such a system may be used for a variety of applications, including high‐resolution image aberration correction, surface metrology, and quantitative phase imaging.

show abstract

“…The reconstruction may be further accelerated via parallel computing or via more advanced algorithms. [ 31 ]…”

Section: Resultsmentioning

confidence: 99%

Single‐Shot Wavefront Sensing with Nonlocal Thin Film Optical Filters

Li,

Jia,

Jin

et al. 2023

Laser & Photonics Reviews

View full text Add to dashboard Cite

show abstract

“…Lately, NNs, and the more modern deep NNs, have also been used for the wavefront reconstruction step (see, e.g., Refs. [35][36][37][38]. The results indicate that NN reconstruction is less sensitive to non-linearity and increases the operational range of the pyramid WFS.…”

Section: Related Workmentioning

confidence: 99%

Laboratory experiments of model-based reinforcement learning for adaptive optics control

Nousiainen,

Engler,

Kasper

et al. 2024

J. Astron. Telesc. Instrum. Syst.

View full text Add to dashboard Cite

Direct imaging of Earth-like exoplanets is one of the most prominent scientific drivers of the next generation of ground-based telescopes. Typically, Earth-like exoplanets are located at small angular separations from their host stars, making their detection difficult. Consequently, the adaptive optics (AO) system's control algorithm must be carefully designed to distinguish the exoplanet from the residual light produced by the host star. A promising avenue of research to improve AO control builds on datadriven control methods, such as reinforcement learning (RL). RL is an active branch of the machine learning research field, where control of a system is learned through interaction with the environment. Thus, RL can be seen as an automated approach to AO control, where its usage is entirely a turnkey operation. In particular, model-based RL has been shown to cope with temporal and misregistration errors. Similarly, it has been demonstrated to adapt to nonlinear wavefront sensing while being efficient in training and execution. In this work, we implement and adapt an RL method called policy optimization for AO (PO4AO) to the GPU-based high-order adaptive optics testbench (GHOST) test bench at ESO headquarters, where we demonstrate a strong performance of the method in a laboratory environment. Our implementation allows the training to be performed parallel to inference, which is crucial for on-sky operation. In particular, we study the predictive and self-calibrating aspects of the method. The new implementation on GHOST running PyTorch introduces only around 700 μs of in addition to hardware, pipeline, and Python interface latency. We open-source well-documented code for the implementation and specify the requirements for the RTC pipeline. We also discuss the important hyperparameters of the method and how they affect the method. Further, the paper discusses the source of the latency and the possible paths for a lower latency implementation.

show abstract

“…The Zernike coefficients are generated randomly based on the Kolmogorov turbulence model [32].The ranges of turbulence model parameter D/r0 are set from 5 to 10 (D is the effective aperture of the SHWFS, and r0 is the coherence length of the atmosphere). A 16 × 16 microlens array with 192 valid subapertures is used to sufficiently sample the incident wavefronts [33]. The key parameters of the SHWFS are list in Table . I.…”

Section: Data Generation and Network Trainingmentioning

confidence: 99%

Centroid-Predicted Deep Neural Network in Shack-Hartmann Sensors

Zhao

Wang

et al. 2022

IEEE Photonics J.

View full text Add to dashboard Cite

The Shack-Hartmann wavefront sensor produces incorrect wavefront measurements when some sub-spots are weak and missing. In this paper, a method is proposed to predict the centroids of these sub-spots for the Shack-Hartmann wavefront sensor based on the deep neural network. Using the centroid information of present sub-spots, the method is able to predict the absent sub-spots' positions. The feasibility and effectiveness of this method are verified by a large number of numerical simulations. The method is applied to wavefront measurement of light with non-uniform near-field intensity. The simulation results show that the proposed method is of great help to improve the measurement accuracy and the Strehl ratio of the focal spot. For wavefronts outside of the training sample, the proposed method shows good generalization and adaptability. In addition, the experiment results demonstrate that the proposed method can predict the missing sub-spots' centroid displacements accurately even though a large proportion of sub-spot is lost randomly.

show abstract

Deep learning wavefront sensing method for Shack-Hartmann sensors with sparse sub-apertures

Cited by 26 publications

References 26 publications

Single‐Shot Wavefront Sensing with Nonlocal Thin Film Optical Filters

Single‐Shot Wavefront Sensing with Nonlocal Thin Film Optical Filters

Laboratory experiments of model-based reinforcement learning for adaptive optics control

Centroid-Predicted Deep Neural Network in Shack-Hartmann Sensors

Contact Info

Product

Resources

About