Step phase reconstruction using an anisotropic total variation regularization method in a diffraction phase...

Area of Science:

• Optics and Photonics
• Microscopy Techniques
• Image Processing

Background:

• Diffraction phase microscopy (DPM) is a powerful technique for label-free imaging.
• Accurate phase retrieval is crucial for quantitative analysis of microscopic samples.
• Estimating phase with sharp features, like step edges, remains a challenge in DPM.

Purpose of the Study:

• To develop and validate a novel approach for phase distribution estimation in DPM.
• To accurately retrieve phase maps containing step features or edge characteristics.
• To demonstrate the method's efficacy for analyzing surface topography.

Main Methods:

• Formulation of a cost function for interferograms in DPM.
• Application of anisotropic total variation regularization-based optimization.
• Minimization using a proximal gradient method with iterative amplitude and phase updates.

Main Results:

• Numerical simulations confirm the approach's effectiveness in retrieving phase maps with rapid edge variations.
• Experimental validation demonstrates accurate surface topography measurement using the proposed method in DPM.
• The method successfully handles complex phase distributions with discontinuities.

Conclusions:

• The proposed method provides a robust solution for phase estimation in DPM, particularly for samples with sharp features.
• This technique enhances the quantitative capabilities of diffraction phase microscopy for surface topography analysis.
• The approach offers improved accuracy and reliability in phase retrieval for microscopic imaging applications.