针对当前单帧条纹技术存在的精度受到影响等问题及传统相位测量轮廓术受多帧采集限制的情况, 提出了一种基于单帧变形条纹的空间等步相移轮廓术。先提取所采集的单帧变形条纹图的零频成分, 通过滤除背景光并归一化处理, 可获得振幅为常数1的纯交流正弦变形条纹, 当该纯交流正弦变形条纹与该条纹顺延若干像素后的正弦变形条纹相加时, 根据和差化积即可获得一幅新的具有一定相移量的纯交流正弦变形条纹, 依此反复即可获得5帧相移变形条纹, 从而可以利用5帧相移解相算法重构待测物体的三维面形。实验验证了本方法的可行性和实用性。与传统相移轮廓术相比, 所提方法仅需要获取一帧变形条纹图, 无需采集多帧变形条纹, 可以实现实时三维测量, 而与傅里叶变换轮廓术相比有更高精度。

For the problem that the accuracy of existed single-shot technologies is affected, and the traditional phase measurement profilometry(PMP) is limited by multi-frame acquisition, a spatial PMP based on single-shot deformed pattern is proposed. First the zero-frequency component of the acquired single-frame deformed pattern is extracted. By background filtering out and normalizing, the AC sinusoidal deformed pattern with an amplitude of constant 1 can be obtained. When the AC sinusoidal deformed pattern is added by the pattern but a certain pixels mismatch, a new pure AC sinusoidal deformed pattern with a certain shift phase can be obtained according to trigonometry operations. In this way, five phase-shifting deformed patterns can be obtained, so the three-dimension shape of the object can be reconstructed with a phase calculation algorithm. The experimental results show that the feasible and validity of the proposed method. Compared with the traditional PMP, the proposed method only needs one deformed pattern and avoid capturing multiple patterns. So it may play a potential application in real-time measurement, and compared with the Fourier transform profilemetry, it shows a higher precision.