针对内窥镜手术监控中二维图像三维重建的需求,对结构光三维重建中的标定和匹配中的关键技术进行改进,使用改进后的标定方法和编码策略有效重建了动物内脏目标。为减少错检,在格雷码条纹之间加入一个像素宽的红色,来判断条纹边界像素点的存在与否。使用灰度重心法可精确实现条纹边界定位。针对三步相移法标定投影仪过程中相位去包络的不稳定性问题,使用双频投影栅线法,避免了复杂易产生噪声的去包络过程。实验结果表明,使用文中提出的策略可以有效重建动物内脏目标。
This paper improves the calibration and matching key technology in terms of 3D reconstruction for a 2D image in a surgery monitored by an endoscope. Animal internal organ is reconstructed effectively by the improved calibration and coding scheme. To induce false detection, a pixel red is added between the Gray code stripes, which decides whether boundary exists nearby, The gray weighted can- troid method is improved to locate the stripe boundary. Aiming at the instability of the phase unwrapping in the process of the projector cali- bration by three-step phase shifting method, a two-frequency grating projeetion method is used to avoid the complex unwrapping process. The experimental results show that the animal internal organ is reconstructed effectively by the proposed scheme.