用于三维傅氏轮廓术的一种自适应基频带通滤波器

Adaptive main frequency bandpass filters used in Fourier transform profilometry

介绍了傅氏轮廓术的基本原理,通过仿真与调相频谱分析,提出了一种新的自适应基频带通滤波器算法。利用此算法设计出了基频带通滤波器并用于傅氏轮廓术的位相提取。结果表明,相对于非自适应性的滤波器,使用本文设计的滤波器其三维面型测量精度可以提高5%,并且具有稳定的重复测量精度,在不受主观因素的影响下实现了实时三维面形重建。以螺纹钢面型为例进行了位相提取实验,实验中计算了面型测量精度,并与理论仿真精度进行比对与分析,得出的实验结果与理论仿真结果相一致,验证了该算法的有效性。

The basic principle of Fourier Transform Profilometry（FTP） is introduced, then a new Adaptive Main Frequency Bandpass Filter（A-MFBF） algorithm is proposed based on the simulation result of phase mod- ulation frequency spectrum. The proposed algorithm is used to design the A-MFBF to filter out the main frequency and reconstruct the phase of a measured object. Results show that the precision of 3-D measurement system can be improved by 5% relative to that of the average ones, and its metrical precision can be repeated steadily. The real-time 3-D shape reconstruction is implemented by proposed A-MFBF without effect by subjective factors. Furthermore, a experiment of reconstructing phase for the steel with ribs is carried out, and the precision of reconstructed measurement is calculated and analyzed. The results show that the precision from the theoretical simulation is in good agreement with the experimental results, which demonstrates the validity of the algorithm.