提出了一种五步非等步相移算法,并成功应用于在线三维测量中。仅投影一固定正弦光栅条纹到物体上,通过物体在线运动产生等效相移,在一个条纹周期范围内任意采集五帧变形条纹图,通过像素匹配使各帧条纹图中的物点一一对应并计算出相应的等效相移量。采用所提出的五步非等步相移算法,即可重构物体。计算机模拟与实验验证了该方法的可行性和有效性,表明非等步相移算法在在线三维测量中更具适应性,有利于提高三维测量精度。

A novel five unequal step length phase-shifting algorithm is proposed. It is applied to on-line 3D measurement successfully. While a static sinusoidal fringe is projected on the measured object, five arbitrary phase shifting deformed patterns are captured by CCD as the measured object moves within a cycle of the fringe, pixels matching can make the object at each frame of deformed patterns be identical point -to-point so as to caculate the corresponding equivalent shifted phase on the deformed patterns, the profile of the measured object can be reconstructed by the proposed algorithm. The computer simulations and experimental results verify its feasibility and validity which show more adaptable and higher precision in on-line 3D measurement.