Isotropic inverse-problem approach for two-dimensional phase unwrapping.

Area of Science:

• Image processing
• Computational imaging
• Optical metrology

Background:

• Phase unwrapping is crucial for quantitative phase imaging.
• Existing methods struggle with complex phase data and rotation.
• Tomographic phase microscopy generates wrapped phase images requiring unwrapping.

Purpose of the Study:

• To introduce a new, rotation-invariant 2D phase unwrapping technique.
• To improve the accuracy and applicability of quantitative phase evaluation.
• To address limitations of current phase unwrapping algorithms.

Main Methods:

• Formulating phase unwrapping as an inverse problem.
• Minimizing an energy functional with weighted data fidelity and higher-order total variation regularization.
• Implementing a rotation-invariant algorithm.

Main Results:

• The proposed method successfully unwraps phase from simulated and real tomographic phase microscope data.
• Demonstrated superior performance compared to state-of-the-art methods, particularly in handling complex phase distributions.
• Achieved rotation invariance, expanding the range of applicable images.

Conclusions:

• The new technique significantly enhances 2D phase unwrapping capabilities.
• It broadens the applicability and outreach of quantitative phase evaluation methods.
• This approach offers a robust solution for challenging phase imaging scenarios.