Ponencia presentada en el XIX Congreso Internacional de Acústica (ICA2007), Madrid, 2-7 Sep 2007.-- PACS: 43.60.Hj.
Structural noise is an important distortion source in ultrasonic images of highly dispersive materials. This noise can considerably mask the flaw signals and cannot be eliminated using standard techniques for noise cancelling like band-pass filtering or temporal averaging.
Fortunately, this noise has several characteristics that may help an algorithm designer to develop methods that improve the defect visibility.
In the literature there exist many techniques for flaw visibility enhancement in the presence of structural noise. However, many of the proposed algorithms do not systematic exploit the above mentioned characteristics. In this work we analyze the structural noise from a point of view of an algorithm developer that seeks parameters that may help to eliminate the noise enhancing the defect visibility. Also, we present several algorithms examples that, using this information, dramatically improve the visibility of the flaw echoes.
The authors would like to thank the financial support of Spanish Ministry of Science and Education with project DHORAVE BIA2006-15188-C03-01. M. Molero is supported by the Programme Alban, the European Union programme of high level Scholarship for Latin America (E05D055458MX) and The Mexican National Council for Science and Technology Conacyt:
(186384). M.G. Hernández is supported by a postdoctoral CSIC-I3P contract granted by the European Social Fund.
Peer reviewed