2-D Continuous Wavelet Transform for ESPI phase-maps denoising

• Our denoising technique does not require the previous orientation estimation, a strong computational effort and a long processing time.
• We proposed a practical implementation of our proposal and we provide the MATLAB function code.
• The use of Gabor atoms can naturally model phase fringes, and has a good performance against noise preserving phase fringes.

In this work we introduce a 2-D Continuous Wavelet Transform (2-D CWT) method for denoising ESPI phase-maps. Multiresolution analysis with 2-D wavelets can provide high directional sensitivity and high anisotropy which are proper characteristics for this task. In particular, the 2-D CWT method using Gabor atoms (Gabor mother wavelets) which can naturally model phase fringes, has a good performance against noise and can preserve phase fringes. We describe the theoretical basis of the proposed technique and show some experimental results with real and simulated ESPI phase-maps. As can be verified the proposal is robust and effective.