The ideal speckle pattern is high contrast, contains random speckles, consistent dot sizes, and equal amounts of black and white on the sample’s surface. If all of these conditions are met, the area of interest will produce low levels of noise and may be tracked with very high levels of certainty.
Below is an example of how various speckle patterns can effect measurement noise. The middle pattern was created using a roller from the VIC Speckle Pattern Application Kit, while the patterns on either side were created using spray paint. As you can see, the roller from the speckle pattern kit produces a high contrast pattern with approximately 50% black and white coverage, while the painted pattern produces a lower contrast pattern with inconsistent dot sizes.
Figure 1:A relatively flat plate with three different speckle patterns analyzed to show the variable Sigma_Z
Figure 1 shows a 2D data overlay of the variable Sigma_Z, which is the confidence margin of the measurements taken in the Z direction (out-of-plane shape measurement). The closer Sigma is to zero, the more statistically likely the value shown is indeed the correct value. The outside patterns are low in contrast and overly fine which result in larger Sigma_Z values, which indicates less confidence. The middle pattern shows lower Sigma_Z values, which indicates high confidence in the measurement. The results speak for themselves.
Figure 2:The 3D plot of Sigma_Z displays how poor patterns create low confidence margins and noisy shape measurements.
High Sigma values in VIC-2D and VIC-3D indicate higher measurement uncertainty, which can result in an increase in measurement noise in the shape, displacement, and strain data. As Figure 2 shows, the outside patterns show a larger range of Z values of a relatively flat sample, while the center pattern shows much smaller bias.
In Figure 3 below data for the variable exx (transverse strain in the X direction) is shown for the same plate after a rigid body translation, where there should be zero strain. The range of strain values shown in the contour plot are much higher on the outside patterns than the center pattern. The result shows that the local strain measurements on the outside patterns are more noisy and will produce data that is less accurate. This is a good illustration of how poor speckle patterns can result in noisy measurements, while favorable patterns (applied with the speckle kit) will produce much better results.
By applying the optimal speckle pattern using the VIC Speckle Pattern Application Kit your results will be less noisy and more accurate.
The VIC Speckle Pattern Application Kit from Correlated Solutions, Inc. can help produce very low noise and very high confidence results for DIC measurements. The speckle kit allows both new and experienced DIC users to create ideal and repeatable patterns for many fields of view for nearly any lens and sensor size combination. Having this capability can streamline the sample preparation process, ensure consistent patterns for recurring tests, and provide a means to obtain a consistently low strain noise floor for all measurements acquired with your VIC-2D or VIC-3D system.