In any automated algorithm for interpreting photoelastic fringe patterns it is necessary to understand and quantify sources of error in the measurement system. We have been considering how the various components of the coating affect the photoelastic measurement, because this source of error has received fairly little attention in the literature. Because the reflective backing is not a perfect retroreflector, it does not preserve the polarization of light and thereby introduces noise into the measurement that depends on the angle of obliqueness and roughness of the reflective surface. This is of particular concern in resolving the stress tensor through the combination of thermoelasticity and photoelasticity where the components are sensitive to errors in the principal angle and difference of the principal stresses. We have developed a physical model that accounts for this and other sources of measurement error to be introduced in a systematic way so that the individual effects on the fringe patterns can be quantified. Simulations show altered photoelastic fringes when backing roughness and oblique incident angles are incorporated into the model.
© 2000 Optical Society of America
(100.5010) Image processing : Pattern recognition
(120.4290) Instrumentation, measurement, and metrology : Nondestructive testing
(240.0240) Optics at surfaces : Optics at surfaces
(260.1440) Physical optics : Birefringence
Deonna Woolard and Mark Hinders, "Model for Quantifying Photoelastic Fringe Degradation by Imperfect Retroreflective Backings," Appl. Opt. 39, 2043-2053 (2000)