Research On Feature Extraction And Simplification Of 3d Point Cloud

With the development of science and technology,3D point cloud data is gradually being applied to fields such as medical care and reverse engineering due to its advantages such as convenience in acquisition and simplicity in application.At the same time,as the accuracy of 3D scanning continues to increase,the amount of acquired 3D point cloud data becomes increasingly large.The resulting data redundancy problem is very disadvantageous to the subsequent point cloud processing.Therefore,the simplification of the point cloud data has become a hot topic in the next study.This dissertation focuses on the feature points extraction of scattered point clouds and the reduction of the higher accuracy of point clouds on the basis of retaining the appropriate feature points.The specific research contents are as follows:Firstly,in the aspect of feature extraction of point cloud,this paper proposed an algorithm of extracting feature points based on multiple criterions,which aimed to extract boundary feature points and sharp feature points efficiently.At the beginning,the algorithm built the point cloud topological structure based on a modified k-d tree approach to search for the K-nearest neighborhood of the sample point;then,according to each K-nearest neighborhood points,it calculated three feature parameters based on vector angle criteria,kernel density criteria and field power criteria;finally,according to the three parameters,it obtained feature discriminant parameters and the global fixed threshold.A point was recognized as the feature point when its value of discriminant parameter is bigger than the threshold.This paper comprehensively considers the advantages of various criterions and makes up for the weakness of the single criteria through the weighted calculations,making the determination of feature points more accurate and comprehensive.Secondly,in the aspect of point cloud simplification,this paper proposed an algorithm of point cloud simplification based on the vector angle fuzzy entropy,which aimed to solve the problems that previous algorithms existed,such as point cloud features could not fully preserved,and the voids were easily caused by simplification in plane or non-characteristic regions.At the beginning,the point cloud feature extraction algorithm was used to extract and preserve the point cloud feature points.Then a reasonable point cloud segmentation threshold could be selected by obtaining the minimum normal angle fuzzy entropy.Finally,the point cloud data was iteratively segmented into multiple subsets,and made use of different sampling rates to reasonably simplify each subset to achieve effective point cloud reduction.Finally,the experiments on multiple point cloud models and comparison with other algorithms proved the effectiveness and accuracy of the proposed algorithm.