Abstract Digital fringe projection (DFP) method has been widely used in three-dimensional (3D) shape measurement. Lots of studies have been conducted to enhance the measurement accuracy and speed by modulating the patterns. However, it is still hard to acquire enough number of 3D results to ensure the continuity of dynamic 3D shape measurement when phase shifting profilometry and temporal phase unwrapping algorithm is adopted. Besides, object motion can introduce phase error and thus measurement error for phase shifting profilometry. To this end, a dynamic 3D reconstruction framework based on modified three-wavelength phase unwrapping algorithm and phase error compensation method is proposed to solve this problem. By redefining the order of fringe patterns projected by the digital light processing (DLP) projector, the number of required patterns for 3D reconstruction can be reduced and the speed of dynamic 3D shape measurement can be improved. And using phase error compensation method based on Hilbert transform, the accuracy of 3D reconstruction can also be improved. Experiments were carried out to demonstrated the effectiveness of the proposed framework. And the modified algorithm successfully improved the speed of dynamic 3D shape measurement by 1 ∕ 3 .

中文翻译：

---

用于动态三维形状测量的改进三波长相位展开算法

摘要 数字条纹投影（DFP）方法已广泛应用于三维（3D）形状测量。已经进行了大量研究以通过调制模式来提高测量精度和速度。然而，当采用相移轮廓和时间相位展开算法时，仍然难以获得足够数量的3D结果来保证动态3D形状测量的连续性。此外，物体运动会引入相位误差，从而导致相移轮廓测量的测量误差。为此，提出了一种基于改进的三波长相位展开算法和相位误差补偿方法的动态3D重建框架来解决这个问题。通过重新定义数字光处理 (DLP) 投影仪投射的条纹图案的顺序，可以减少 3D 重建所需的图案数量，提高动态 3D 形状测量的速度。并且使用基于希尔伯特变换的相位误差补偿方法，也可以提高3D重建的精度。进行了实验以证明所提出框架的有效性。并且改进后的算法成功地将动态 3D 形状测量的速度提高了 1 ∕ 3 。