Mitigating the ambiguity problem in the CNN-based wavefront correction.

Area of Science:

• Optics and Photonics
• Computational Imaging
• Artificial Intelligence in Optics

Background:

• Phase retrieval in adaptive optics (AO) systems is crucial for correcting wavefront aberrations.
• Traditional methods often suffer from ambiguity issues, limiting their effectiveness.
• Convolutional Neural Networks (CNNs) show promise for wavefront sensing and correction.

Purpose of the Study:

• To develop an advanced attention-based adaptive optics method for improved phase retrieval.
• To address the phase ambiguity problem inherent in traditional phase retrieval techniques.
• To enhance wavefront correction performance compared to existing CNN-based methods.

Main Methods:

• Integration of phase diversity with a CNN using a non-local attention block.
• Development of an attention-based mechanism for enhanced feature extraction.
• Utilizing simulation environments to validate the proposed method.

Main Results:

• The proposed attention-based method successfully mitigates the phase ambiguity problem.
• Demonstrated superior performance in wavefront correction over traditional CNN-based approaches.
• Simulation results confirm the effectiveness and robustness of the novel technique.

Conclusions:

• The attention-based adaptive optics method offers a significant advancement in phase retrieval.
• This approach provides a more accurate and reliable solution for wavefront correction in optical systems.
• The integration of non-local blocks enhances the capability of CNNs in complex AO applications.