首页|Error suppression for speckle correlation-based 3D shape measurement via dynamic shape function selection
Error suppression for speckle correlation-based 3D shape measurement via dynamic shape function selection
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NSTL
Elsevier
? 2021Speckle correlation-based 3D shape measurement technology has an increasing number of applications in the fields of medical assistance, cultural relic protection, product manufacturing, and inspection. The foremost process of this method is the stereo matching of speckle image pairs using digital image correlation (DIC). It is challenging to improve the matching precision of DIC for complex shapes with large or abrupt curvature changes. Because the local curvature change and surface shape complexity of different objects vary, first-order and second-order shape functions have different abilities to describe the mapping relationship from the reference subset to the target subset. As a result, the systematic errors caused by subset under-matching and over-matching are also different. In this study, a pixel-by-pixel dynamic shape function selection algorithm is proposed for different measurement objects or different regions on the surface of the same object. Experimental results show that the proposed algorithm can effectively suppress the measurement error caused by the limited description ability of a specific shape function.
NSTL
Elsevier
