Anisotropic phase-map denoising using a regularized cost-function with complex-valued Markov-random-fields
In our recently reported work [1] (Villa et al., 2009) we derived a regularized quadratic-cost function, which includes fringe orientation information, for denosing fringe pattern images. In this work we adopt such idea for denoising wrapped phase-maps. We use a regularized cost-function that uses c...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2010 |
| País: | España |
| Institución: | Universidad Complutense de Madrid (UCM) |
| Repositorio: | Docta Complutense |
| Idioma: | inglés |
| OAI Identifier: | oai:docta.ucm.es:20.500.14352/43967 |
| Acceso en línea: | https://hdl.handle.net/20.500.14352/43967 |
| Access Level: | acceso abierto |
| Palabra clave: | 535 Fringe-Pattern-Analysis Fourier-Transform Interferometry Images Filter Radar Óptica (Física) 2209.19 Óptica Física |
| Sumario: | In our recently reported work [1] (Villa et al., 2009) we derived a regularized quadratic-cost function, which includes fringe orientation information, for denosing fringe pattern images. In this work we adopt such idea for denoising wrapped phase-maps. We use a regularized cost-function that uses complex-valued Markov random fields (CMRFs) with orientation information of the filtering direction along isophase lines. The advantage of using an anisotropic filter along isophase lines is that phase and noise can be properly separated while 2 pi phase jumps are preserved even in high frequency zones. Apart from its robustness, the outstanding advantage of our method is its minimal computational effort. We present some results processing simulated and real phase-maps. |
|---|