基于机器视觉的镜面面形检测算法研究与应用

Research and Application of Morror Surface Topography Measurement Algorithm Based on Machine Vision

针对大面积镜面面形测量,提出了基于机器视觉与镜面条纹反射的面形精密测量、三维重建和面形拼接算法,并设计了面形检测系统。该系统采用相机拍摄定日镜反射的条纹图像,通过四步相移法对相位进行解析得到不同相位下反射后的条纹图像;基于时间解包裹运算得到的相位差图实现对镜面面形的三维重建。搭建了镜面面形检测系统,验证表明:所提出的面形检测算法和面形测量系统与设计要求相符,对大面积镜面面形的高效测量具有重要的应用价值。

For large-area mirror surface measurement,an algorithm including data stitching for a precision gauge of specular surface topography measurement and 3-D shape reconstruction based on machine vision and fringe reflection was proposed,a surface shape detection and verification system was designed.In this system,the fringe images reflected by heliostat are taken by camera,and the fringe images reflected by different phases are obtained by four step phase-shifting method.Based on the phase difference map obtained by time unwrapping operation,the three-dimensional reconstruction of the mirror surface is realized.The test results show that the proposed algorithm and system are in accordance with the design requirements,and it has important application value for the efficient measurement of large area mirror surface.