Convex optimization-based blind deconvolution for images taken with coherent illumination.

Area of Science:

• Optics and Image Processing
• Convex Optimization

Background:

• Blind deconvolution is crucial for image restoration.
• Coherent illumination presents unique challenges for deconvolution.

Purpose of the Study:

• To propose a rank-constrained reformulation for blind deconvolution.
• To develop a novel convex heuristic for this problem.

Main Methods:

• Reformulating blind deconvolution with a rank constraint.
• Developing a convex heuristic incorporating prior information and phase diversity.
• Iterative re-parameterization for refining estimates.

Main Results:

• The proposed heuristic effectively handles rank constraints in blind deconvolution.
• Incorporation of prior information and phase diversity enhances performance.
• Iterative re-parameterization leads to improved estimation accuracy.

Conclusions:

• The novel convex heuristic offers a robust approach to blind deconvolution under coherent illumination.
• The method demonstrates potential for enhanced image restoration in relevant applications.