结构光数字栅线法对物体三维形貌的高精度测量

为了对物体的微观三维形貌进行高精度测量，开发了基于结构光的数字栅线三维纹理测量系统，该系统利用解位相的方法进行物体三维高度信息的计算。首先对相位主值的计算，采用了4步等步长相移法采集16幅等步长的相移图像，分成4组分别计算相位主值，再求其平均值作为最终的相位主值。为了滤除噪声及投影仪、CCD相机畸变等各种影响，对采集的条纹图像采用了自适应维纳滤波和小波多阈值方法进行预处理;其次对于位相展开，提出了基于多频的梯度导向相位展开算法，能够有效避开相位间断点，具有稳定的展开结果。实验结果表明，此系统所使用的方法在条纹图的P₀=22.7 mm-1时，获取的三维信息分辨率为2.75 μm，高度信息精度优于0.5 μm，同时具有测量快速，操作简单，无接触等特点，能够满足高精度的微观三维形貌测量的要求。

High-Precision 3D Morphology Measurement by Digital Gatling Method Based on Structured Light

In order to measure the microscopic 3D morphology of the objects with high-precision, a 3D texture measurement system of digital gatling based on structured light was designed, which can calculate the 3D height information with the analytic phase method. First, the authors collected sixteen equal step phase images by the four-step equal step method, and calculated their main value by dividing them into four groups. Then，the authors found the average as the final phase main value. The pretreatment on the fringe was done by the adaptive Wiener filter and wavelet multi-threshold method to eliminate the various effects of noise, projector distortion and CCD camera distortion. Besides, gradient-oriented phase unwrapping algorithm based on multi-frequency was introduced to avoid phase discontinuity point in the course phase unwrapping, and it was proven to be effective and stable. Experiments showed that the system’s 3D resolution was 2.75 μm, and the high degree accuracy was better than 0.5 μm, when the system was running with the fringe parameter p₀=22.7 mm-1. In addition, the system has many advantages such as fast measuring, simple operation and non-contact, which can meet the need of the high precision measurement requirements for the microscopic 3D morphology.