Speckle beam sensing via quadriwave lateral shearing interferometry assisted by convolutional neural networks

Area of Science:

• Optics and Photonics
• Artificial Intelligence in Imaging
• Wavefront Sensing

Background:

• Quadriwave lateral shearing interferometry (QLSI) is a robust quantitative phase imaging technique.
• Traditional QLSI struggles with speckle beams containing phase singularities due to integration errors.

Purpose of the Study:

• To develop a novel method for accurate speckle beam reconstruction using QLSI.
• To overcome the limitations of phase gradient integration in QLSI for complex wavefronts.

Main Methods:

• A convolutional neural network (CNN) was trained to directly retrieve complex amplitude from QLSI interferograms.
• The CNN bypasses traditional phase gradient integration for improved accuracy.

Main Results:

• The CNN-assisted QLSI method accurately reconstructed amplitude and phase profiles of complex speckle beams.
• Achieved a Pearson correlation coefficient > 0.9 compared to ground truth.
• Demonstrated robustness against phase singularities.

Conclusions:

• The proposed CNN-assisted QLSI method significantly enhances speckle beam reconstruction accuracy.
• This technique is expected to broaden QLSI applications in challenging optical environments.
• Enables advanced optical imaging and free-space laser communication in turbulent conditions.