基于结构光辅助的网格候选点三维测量方法

光学三维测量技术是计算机视觉领域中最为活跃的研究主题之一.针对双目立体视觉三维测量方法中无纹理物体表面测量误差较大的问题,提出一种基于结构光投影辅助的网格候选点三维测量方法.设计灰度单调变化的锯齿波图案投射在被测场景中,有助于使三维网格候选点投影像点间的灰度差异最大化,依此提高系统的测量精度,实现了主动测量方法和被动测量方法的有效结合.通过3DS MAX和MATLAB软件平台下的仿真实验,验证了结构光辅助的网格候选点测量方法在计算精度上具有明显优势,且不会增加系统的复杂度,扩展了网格候选点三维测量方法的适用范围.

Three-dimensional measurement method based on mesh candidates assisted with structured light

Optical three-dimensional measurement technology is one of the most active research topics in computer vision domain. Aiming at the problem of larger measurement error for the surface without texture, a mesh candidate based method with structured light illuminating is introduced for adding texture on the scene via further research of the preparatory work. The illuminating sawtooth pattern projected from the mesh candidates in the scene makes the similarity of the corresponding points more distinguishable. The active and passive optical measuring methods are combined effectively. Simulation experiments on 3DS MAX and MATLAB software platforms prove that the precision of the system has been improved obviously.