基于ASODVS的全景点云数据获取技术的研究

Data Acquisition of Point Cloud Based on ASODVS

摘要:

针对现有三维激光扫描设备不能同时获取空间点云的坐标信息和色彩信息,多视角点云数据配准复杂等问题,采用了一种将全方位视觉传感器ODVS(Omni-directional Vision Sensor)和可移动360°面激光光源相结合的主动式全景视觉传感器ASODVS(Active Stereo Omni-directional Vision Sensor)来获取点云数据。通过人机接口对ASODVS的扫描步长和速度等进行控制,实现了边扫描边获取全景点云数据;对获取点云数据空间信息过程中的关键技术——激光投射点提取算法进行了深入研究;采用了3种不同的全景激光线提取算法对全景切片图像中激光投射点进行提取,并实验研究了各自的优劣;实验结果表明,对于ASODVS而言,帧间差提取算法能有效快速准确的提取三维全景场景中激光投射点,并具有消耗计算机资源少、操作过程自动化、生成数据有序等优点。

Abstract:

It is difficult to acquire surface geometry and color data of space point cloud simultaneously for the existing 3D laser scanners, and matching the point cloud data among different viewing angles is also hard. This paper is aimed to use an ASODVS(Active Stereo Omni-directional Vision Sensor) made up by a type of ODVS(Omni-directional Vision Sensor) and 360°mobile laser source to solve these problems. Point cloud data can be scanned and acquired at the same time using ASODVS whose scanning step and speed are controlled by a man-machine interface. Algorithm of extracting laser line which is the key technology of point cloud data acquisition is researched deeply in this paper. Three different algorithms of extracting laser line from panorama images were used in this paper to compare with each other. Experimental results show that, the algorithm of frame difference can extract laser project points in panorama images accurately and efficiently. The algorithm has advantages in less computer resources consuming, process automatically and generating data orderly.