大视场下线结构光光条中心的快速提取

Rapid extraction of line-structured light stripe in large field of view

在线结构光三维测量中,线结构光光条中心的提取是关键的一步。针对强背景光大视场下线结构光光条(长度约2m)中心的提取,提出了一种基于感兴趣区域(ROI)的光条中心提取方法。首先用统计的方法确定图像处理的ROI,再在此区域内用最大类间方差法进行阈值分割并用灰度重心法提取中心。结果表明:在VisualC++6.0平台上,处理一帧1280×1024大小的线结构光条图像大约用时47ms,且光条中心提取精度高。这种提取中心的方法很好地减弱了大视场下强背景光的干扰,提高了光条中心提取速度。

In three-dimensional testing systems based on line-structured light,extracting the center of a line-structured light stripe quickly and accurately is a key problem. To extract the center of line-structured light with a length of 2 meter in a large field of view and strong noise background,an exaction method based on region of interest （ROI） is proposed. Firstly,ROI for image processing is defined by using statistic method,then Otsu method is used for threshold segmentation and barycenter method is used to exact the center within ROI. Experiments showed that the extracting time of a 1280×1024 sized picture of a line-structured light was about 47ms in Visual C ＋＋6.0 by the proposed method and the extraction was proved accurate. The proposed method reduces the interference of strong background noise in large field of view and improves the extraction speed of line-structured light.