ABSTRACT
One of the major difficulties of reconstructing three-dimensional (3D) shapes from stereo vision is the correspondence problem. Speckle can help in this aspect by providing unique patterns for area-based matching. After applying a high-contrast random speckle pattern on the object surface, disparities map with accuracy up to 0.02 pixels and ultrahigh spatial resolution can be established. The disparity information can then be used for accurate 3D reconstruction by triangulation. This chapter will introduce various algorithms of digital image correlation (DIC) for speckle pattern matching and various algorithms for 3D shape reconstruction from the disparity map. Examples of 3D shape measurement in various fields will also be given.
