多路径模型对相位测量轮廓术的影响

Effect of Multi-path Model for Phase Measuring Profilometry

为了克服多路径效应对相位测量轮廓术的影响,提出一种相位融合算法.首先建立多路径效应的数学模型,并利用该模型分析多路径效应引起的相位误差,得出了该效应引起的相位误差主要发生在反射率变化剧烈的地方,其除了与邻域内每点的真相位值和真反射率有关外,还与扫描的方向有关的结论;然后基于该分析推导出相位融合算法,沿X,Y方向扫描分别得到2组相位值,再根据物体表面反射率的变化对2组相位进行融合,完成相位的修正.实验结果表明,即使投射大量的编码图片,由多路径效应引入的相位误差仍然存在且不可忽略,采用文中算法可以有效地使相位误差减小2倍.

In order to overcome multi-path effect for phase measuring profilometry, a phase fusion algorithm is proposed. First, we build a mathematical model for the multi-path effect, and then we analyze the multi-path-effect-caused phase error via the model. Second, we find that the phase error caused by the multi-path effect is mainly raised in the area where the reflectivity dramatically varies; and for a pixel, besides the true phase values and the true reflectivity of the pixel＇s neighbors, the multi-path effect is also with regard to scanning direction. Finally, a phase fusion algorithm is derived based on the previous analysis, i. e. , two groups of phase values are obtained by scanning a target along X direction and Y direction respectively, and then the two groups of phase values are fused by employing the corresponding reflectivity of the target. Thus the phase error is corrected. The experimental results demonstrate that such kind of phase error always exists and is not ignorable even if the number of projected patterns is increased, and, with the proposed method, phase error can be effectively reduced by 2 times.