Robust particle-Kalman filtering phase unwrapping algorithm for wrapped fringe patterns.

Area of Science:

• Optics and Photonics
• Signal Processing
• Computational Imaging

Background:

• Phase unwrapping is crucial for interpreting fringe patterns in optical metrology.
• Existing methods struggle with noisy data and accuracy limitations.
• Robust algorithms are needed for reliable phase retrieval.

Purpose of the Study:

• To develop a robust phase unwrapping algorithm for wrapped fringe patterns.
• To improve accuracy and reliability in noisy conditions.
• To formulate phase unwrapping as an optimal state estimation problem.

Main Methods:

• A particle-Kalman filter combining particle and extended Kalman filters.
• Second-order Taylor expansion for state space equations.
• Modified matrix pencil model for phase gradient estimation.
• Quality-guided strategy for path selection.

Main Results:

• The particle-Kalman filter effectively unwraps noisy wrapped fringe patterns.
• Demonstrated improved accuracy and robustness on synthetic and experimental data.
• Outperformed commonly used phase unwrapping methods.

Conclusions:

• The proposed particle-Kalman filter offers a robust and accurate solution for phase unwrapping.
• This method enhances the interpretation of fringe patterns in challenging conditions.
• It provides a significant advancement in optical metrology and signal processing.