WFF-BM3D : a hybrid denoising scheme for fringe patterns

• Main Authors: Zhao, Ming, Qian, Kemao
• Other Authors: Asundi, Anand K.
• Format: Conference or Workshop Item
• Language: English
• Published: 2018
• Subjects: Fringe Pattern; Image Denoising; DRNTU::Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/89570; http://hdl.handle.net/10220/47097
• Institution: Nanyang Technological University

Fringe patterns are widely used in the optical interferometry. The captured fringe patterns usually have noises, and thus a filtering process for fringe patterns is necessary. The windowed Fourier filtering (WFF) algorithm is proposed specially for fringe pattern denoising. The WFF performs well in continuous areas, but has problems in discontinuous areas. In this paper, we proposed a hybrid denoising scheme which combines the advantages of the local WFF algorithm and non-local block-matching 3D filtering (BM3D) algorithm. This new scheme can recover fringe pattern information from the WFF residuals, and improve the denoising results effectively.