线结构光点云粗拼接方法研究

Coarse stitching of point structured cloud of line structured light research on unconstrained method of surface measurement

针对点云拼接需借助点云信息几何特征的问题,基于结构光测量系统模型,提出一种通过提取图像站位信息的方法,解算两图像站位下点云信息的相对位置关系来完成点云的粗拼接。首先,介绍了结构光测量系统的模型,并采用平面靶标的方法对测量系统中的摄像机与激光器的光平面进行精确标定;然后,对光条信息的提取和无约束点云拼接算法进行了研究,通过对二维图像特征点的处理计算出测量系统不同站位下的坐标变换关系;实验结果表明,该方法可对物体表面信息进行有效测量,可将误差降至6.052 mm。为点云精拼接技术提供了支持,并为点云拼接技术在逆向工程检测、三维形貌恢复及目标识别等方面的应用提供了理论依据。

Aiming at the problem of point clouds registration with the help of point clouds information geometric features,a method of extracting image station location information is proposed to accomplish point clouds coarse registration by solving the relative position relationship of point clouds information under two images station based on the structured light measurement system model. First,the model of the structured light measurement system is introduced,and the plane of target is used to accurately calibrate the light plane of the camera and laser in the measurement system. Then,the extraction of light bar information and the unconstrained point clouds splicing algorithm are studied. The coordinate transformation relationship of different stations of the measurement system is calculated by processing the feature points of the two-dimensional image. Finally,the experimental result shows that this method can be effectively applied to the object for accurate measurement,and the error is reduced to 6. 052 mm. It provides support for point clouds fine registration technology,and provides a theoretical basis for the application of point clouds splicing technology in reverse engineering detection,three-dimensional topography recovery and target recognition.