CSI APPLICATION NOTE AN-1701
SPECKLE PATTERN FUNDAMENTALS
INTRODUCTION
In digital image correlation, using an optimal speckle pattern is one of the most important factors in reducing measurement
noise and improving overall results. Understanding the requirements of an ideal speckle pattern and how to apply one to a
specimen facilitates the use of DIC. Here we discuss why a good pattern is needed, pattern requirements, common
application methods, and guidance on some difficult or specific cases.
PATTERN REQUIREMENTS
In DIC, a mesh of small subsets of the image are tracked as the specimen moves and deforms. To perform this tracking, the
subsets are shifted until the pattern in the deformed image matches the reference image as closely as possible; this match
is calculated by the total difference in grey levels at each point.
To achieve effective correlation, our pattern must be
• High contrast: Either dark black dots on a bright white background or bright white dots on a dark black background.
• 50% coverage: We want about equal amounts of white and black on the surface. For example, if the speckles are 5
pixels in size, they should be approximately 5 pixels apart from one another.
• Consistent speckle sizes: speckles should be ideally 3-5 pixels in size in order to optimize spatial resolution, but the
most important thing is that the speckles are consistent in size and not too small (less than 3 pixels in size is too small
and can cause aliased results).
To Next Page
Loading...
Need help?
General
