Outlier detection and robust normal-curvature estimation in mobile laser scanning 3D point cloud data
作者：
Highlights：
• Two statistical techniques are proposed for outlier detection in point cloud data.
• The proposed methods can fit robust plane in laser scanning data.
• The proposed methods produce robust normal and curvature in point cloud processing.
• They are faster and robust than RANSAC, robust PCA and other existing efficient methods.
• They have potential for point cloud denoising, segmentation, and reconstruction.
摘要
Highlights•Two statistical techniques are proposed for outlier detection in point cloud data.•The proposed methods can fit robust plane in laser scanning data.•The proposed methods produce robust normal and curvature in point cloud processing.•They are faster and robust than RANSAC, robust PCA and other existing efficient methods.•They have potential for point cloud denoising, segmentation, and reconstruction.
论文关键词：Feature extraction,Plane fitting,Point cloud denoising,Robust saliency feature,Segmentation,Surface reconstruction
论文评审过程：Received 27 February 2014, Revised 8 June 2014, Accepted 9 October 2014, Available online 22 October 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.10.014
