远距离目标的多基线立体视觉点云成像技术

Point cloud imaging technique for long-distance target using multi-baseline stereo vision

双目立体视觉是机器视觉的重要分支之一,其通过直接模拟人眼观察和图像处理的方式来获取探测目标的深度信息。这种非接触式的测量方法具有速度快、精度高、操作简单等特点,但是对于不同距离的目标,特别是远距离目标难以获取深度信息。提出了一种基于多基线双目立体视觉的远距离目标三维点云成像技术,该方法通过可变基线实现对不同距离目标的点云成像。实验结果表明,该方法能有效探测500 m^4 000 m范围内目标的深度信息,实现三维点云成像,突破了双目立体视觉远距离三维成像的技术难点,同时形成的三维点云与激光三维成像雷达的数据格式一致,为未来激光三维成像雷达技术与立体视觉技术的点云融合,优势互补提供了借鉴。

Binocular stereo vision is one of the most important branches of machine vision,which acquires depth information about the detected target by directly simulating the observation and processing of the image by the human eye. This non-contact measurement method holds the characteristics of high speed,high precision,and simple operation,but it is difficult to obtain depth information for targets at different distances,especially for long distance targets. In this paper,a 3 D point cloud imaging technique for long distance target based on multiple baseline binocular stereo vision is proposed. This method can realize point cloud imaging for targets at different distances by sets of varying baselines. The experimental results demonstrate that the method can effectively acquire the depth information of targets in the range of 500 ~ 4 000 m and realize 3D point cloud imaging,which breaks through the technical difficulties in binocular stereo vision for 3D imaging at long distance. The data format of the generated 3D point cloud is consistent with that of 3D imaging lidar,which serves as a reference for the future point cloud fusion of 3D imaging lidar technology and stereo vision technology.