中文摘要：

建立了基于市售数码相机和投影仪的结构光三维测量系统，研究了系统的数学模型、标定方法及镜头畸变校正方法。在射影几何矩阵模型和三角法解析模型的基础上，结合市售设备内部结构参数精度高、工作稳定的特点，建立了实用数学模型。提出一种借助数码相机赋予投影仪“视觉”功能，向标定模板投射正交格雷码图案，从而确定数字微镜（DMD）像面点和空间标准点对应关系的投影仪标定方法。最后，作为上述数学模型的基础，设计了基于共线特征的校正模板，采用图像法将数码相机镜头畸变参数分离于系统参数之外单独标定。实验结果表明：仿真系统测量相对误差约为0．2％；实际系统测量相对误差〈O．7％。结果验证了方法的可行性，重构的复杂表面具有良好视觉效果。

英文摘要：

In order to develop a structured light 3D measurement system based on off-the-shelf digital camera and projector, a system mathematical model,a calibration scheme and a lens distortion correction scheme were investigated. According to the characteristics of off-the-shelf components such as high precision of internal structure parameters and steady operation, a utility mathematical model based on projection geometry matrix model and trigonometry analytical model was established. A callbration scheme for projector was presented, in which the corresponding relation between Digital Micromirror Device （DMD） elementary points and space standard points was determined in the presence of that the digital camera could endue projector a vision function and could project a orthogonal Gray code pattern onto the calibration board. Finally, the correction board based on eollinear characteristic was designed, and the camera lens distortion parameter separated from system parameters was calibrated separately by image method. The experimental results indicate that relative measurement error of sim- ulation system is about 0.20% and that of real system is less than 0.70%. The feasibility of proposed scheme shows that the complex surfaces reconstructed by real system have a good visual effect.