基于非线性最小二乘算法的点云坐标转换研究

三维激光扫描作为测绘领域的一项高新技术，具有高效率、高精度等优点，但其获取的数据是相对于系统内部坐标系存在的．通过试验，获取某一高边坡的多站扫描数据，同时使用无协作目标全站仪测量站点和标靶的空间坐标，在MATLAB中编程，实现非线性最小二乘算法，完成了扫描坐标向空间坐标的大欧拉角转换．根据转换后两种坐标的较差，近似求取转换的点位误差．可见，该方法对坐标转换具有一定的适用性．点云数据标靶中心的精确提取和全站仪对标靶中心的精确测量是完成坐标精确转换的关键．

Point cloud coordinate transformation based on nonlinear least squares algorithm

As a high-technology in a surveying and mapping field,3D laser scanning has the advantages of high efficiency and high precision.However,the data acquired by the scanning are still related to the instrument coordinate system.The aim of this paper is to find multi-station scanning data of one high-slope,to simultaneously survey the space rectangular coordinates of targets with none cooperation target total station by experiment.Through nonlinear least squares algorithm in MATLAB,the Big Euler Angle coordinate transformation from a scanner coordinate system to a space coordinate system is completed.According to the difference between the two kinds of coordinates,the point error is attained approximately.It is learned that this method is applicable to coordinate transformation,accurately extracting the target center from the point cloud,and that surveying target center with total station is the key to achieve the coordinate transformation accurately.