三维激光扫描拟合直线自动提取算法研究

RESEARCH ON FITTING LINE AUTOMATIC EXTRACTION ALGORITHM OF 3D LASER SCANNING

由于扫描点云数据中存在大量的粗差扫描点,故很多点云数据处理算法都具有较强的抗差性能。但是这些算法为了抵御粗差的影响往往需要进行大量的模型检验计算,导致计算效率较低。利用扫描线数据自身固有的特点,提出一种改进的RANSAC算法。该算法既具有较好的抗差能力,又在计算效率上较传统的抗差算法有较大的提高。利用该算法得到的结果更加准确,扫描点的分割归属更加合理,为后续的拟合平面提取计算提供了更加可靠的基础数据。

Because of the existence of a large robust algorithms have been adopted to processing large amount of modules firstly, and then identify amount of points with gross error in scanned point cloud, many range images data. These algorithms often randomly establish a them individually, so this course is very inefficient. This paper presents an improved RANCAC algorithm which fully utilizes the intrinsic attributions of scanned line data. The algorithm is both efficient and robust. More rational segmentation results can be obtained using the proposed algorithm.