Imaging through strong turbulence with a light field approach

Area of Science:

• Optical Engineering
• Image Processing
• Turbulence Mitigation

Background:

• Strong turbulence severely distorts images, rendering conventional restoration algorithms ineffective due to lost references.
• Existing methods struggle with severe atmospheric distortions, limiting observational capabilities.

Purpose of the Study:

• To develop a novel method using plenoptic sensing for image acquisition and restoration under strong turbulence.
• To identify and utilize moments of minimal wavefront distortion for clearer object imaging.

Main Methods:

• Utilizing a plenoptic sensor (light field camera) to map images onto a cell image array.
• Analyzing sub-aperture images to assess wavefront distortion across the lens aperture.
• Employing a modified Laplacian metric to identify optimal 'good' cell images within a sequence.

Main Results:

• Demonstrated experimental effectiveness of the plenoptic approach under strong turbulence conditions.
• Successfully identified moments and sub-aperture areas with relatively flat wavefront distortion.
• Obtained clearer object representations, revealing details obscured by turbulence in conventional imaging.

Conclusions:

• The plenoptic sensor approach effectively mitigates severe turbulence distortions.
• This method provides a valuable reference for conventional image restoration techniques.
• The system can independently perform object recognition tasks in highly turbulent environments.