Radial phase variation computing: a tool to improve flaw detection in optical diagnosis by shearographic images.

Shearography is an optical and nondestructive technique that has been largely used for damage detection in layered composite materials where delaminations and debondings are found to be among the most common flaws. Shearography detects derivative of the displacements. It is a relative measurement in which two images are recorded for different loading conditions of the sample. The applied loading induces some deformations into the sample, generating a displacement field on its surface. Thermal, acoustical, or mechanical loading are typical excitations applied in a static or dynamic way. The absolute difference between two phase maps recorded at two different loading instances produces an interference fringe pattern, which is directly correlated to the displacements produced on the material surface. In some cases, depending on the loading level and mainly on the sample geometry, interference patterns will contain fringes resulting from geometry changes. This will mask those fringes correlated to flaws introduced into the material, resulting in an image misinterpretation. This phenomenon takes place mainly when the sample has curved geometries, as in, for example, pipe or vessel surfaces. This paper presents an algorithm that uses a mathematical process to improve the visualization of flaws in shearographic images. The mathematical process is based on the calculation of the phase variation, and it is used to search for local deformations contained in the image. This algorithm highlights defect regions and eliminates fringes caused by geometry changes, providing an easier interpretation for complex shearographic images. This paper also shows the principle and the algorithm used for the process. Results, advantages, and difficulties of the method are presented and discussed by using simulated fringe maps as well as real ones.