【摘要】 提出了一种基于包容直线评定不连续平面的方法。首先，基于主成分分析对点云数据进行全局噪声剔除；然后，采用最小二乘平面拟合算法、多阈值点云提取算法及平行直线拟合算法获取符合条件的包容直线，并以空间中2条平行的包容直线代替包容平面建立最小区域；最后实现不连续平面的平面度误差评定及指导平面调整。实验结果表明：该方法与传统的特征点法相比平面度误差值比遗传算法降低0.011 mm,并且根据输出的偏距信息对平面进行调整后，平面度误差值减小1.959 mm。能有效地评定不连续平面的平面度误差和指导平面调整。更多还原

【Abstract】 A method for evaluating discontinuous planes based on inclusive lines is proposed. Firstly, the global noise of point cloud data is eliminated based on principal component analysis. Then, the least square plane fitting method, the multi-threshold point cloud extraction algorithm and the parallel line fitting method were used to obtain the qualified inclusive lines, and two parallel inclusive lines in the space were used to replace the inclusive plane to establish the minimum region. Finally, the flatness error evaluation of discontinuous plane is realized and the plane adjustment is guided. The experimental results show that the flatness error of the proposed method is 0.011 mm lower than that of the genetic algorithm compared with the traditional feature point method, and the flatness error is reduced by 1.959 mm after adjusting the plane according to the output deviation information. It can effectively evaluate the flatness error of discontinuous plane and guide the plane adjustment.更多还原