摘要
如何获得蕴含了深度信息的深度图像是发展机器视觉的关键。因为结构光空间编码方法只需要一幅图像就可以获得被测物体上可测点的三维信息,具有分辨率高、速度快和易于实现的优点而成为研究热点。一种颜色编码方案被用于大型物体表面的三维测量。这种编码方案以红、绿、蓝三原色以及它们的补色黄、品、青6种颜色作为基本颜色,按着任意相邻的3个条纹颜色排列顺序是唯一的方式组合,构成了颜色编码图案,并由投影仪投射到被测物体上。通过对摄像机拍摄的图像进行解码,获得物体的三维坐标。利用被测物体上的标志特征点,对物体进行分块测量。结合边缘检测及圆心检测确定分割线,对图像进行分割、拼接,实现了对大型物体表面的三维测量。
To obtain 3D map including depth information, is one of the key theory and technology in machine vision field. We focused on structured light spatial encoding method, because for 3D information acquiration of the tested object, only one image is needed to be captured by the camera by this method, so it has some advantages such as high resolution, easy implementation and fast measurement. A kind of new method of color encoding is adopted to measure the large object surface. The color stripe encoded pattern w...
出处
《红外与激光工程》
EI
CSCD
北大核心
2008年第S1期162-165,共4页
Infrared and Laser Engineering
关键词
颜色编码
三维测量
大型物体
圆心检测
Color encoding
3D measurement
Large object
Center of circle detection