A twothreshold point cloud feature extraction method with neighborhood adaptive
Zhou Jianzhao, Yan Yuji, Chen Chen, Du Wenchao
(College of Field Engineering,PLA Army Engineering University, Nanjing 210007,China)
Abstract: The feature extraction of point cloud data is an important part of point cloud data processing,which plays a key role in geometric analysis,data segmentation,point cloud registration,and model reconstruction.The point cloud feature extraction technology based on normal vector and curvature has been studied.The problems of neighborhood selection and single parameter calculation in the feature extraction process are clarified.A twothreshold point cloud feature extraction method with adaptive neighborhood is proposed.The experiment compares the extraction effect of the algorithm with the curvaturebased feature extraction algorithm,and verifies the stability and accuracy of the algorithm.This algorithm has a good extraction effect for point clouds with complex geometric features,and has important significance for improving the accuracy and efficiency of point cloud feature point extraction.
Key words : point cloud feature extraction;normal vector;curvature;double threshold;neighborhood adaptive