Robust phase-unwrapping method for phase images with high noise content

Area of Science:

• Optics and Photonics
• Image Processing
• Metrology

Background:

• Phase unwrapping is crucial for quantitative phase imaging, but is challenged by noise and complex fringe patterns.
• Existing methods often struggle with degraded fringe gradients, limiting their applicability in real-world scenarios.

Purpose of the Study:

• To develop a robust phase unwrapping technique capable of handling noisy phase images.
• To address limitations of current methods in the presence of arbitrary fringe patterns and gradient degradation.

Main Methods:

• A novel phase unwrapping approach is presented, focusing on identifying distinct regions between fringe boundaries.
• Phase shifting of identified regions by multiples of 2π is employed for phase reconstruction.
• Image pixels are segmented using a least-squares plane model fit to overlapping domains for robust boundary detection.

Main Results:

• The proposed method demonstrates robustness against fringe gradient degradation.
• It effectively handles noise, filtering artifacts, and limitations from finite instrumentation bandwidth.
• Successful phase unwrapping is achieved on images with arbitrary fringe patterns.

Conclusions:

• The developed phase unwrapping method offers a reliable solution for noisy and complex phase imaging applications.
• Its tolerance to various image imperfections makes it suitable for diverse scientific and industrial metrology tasks.