Flower Feature Extraction And Segmentation Reconstruction Based On 3D Point Cloud

With the development of science and technology,the field of three-dimensional scanning technology is also being deeply studied.At present,3D point cloud data processing technology is applied in various fields,such as virtual reality technology,reverse engineering CAM technology,artificial intelligence and digital city.Virtual plant models are constantly applied in the fields of cultural life,such as digital city,game scene fiction,etc.As an indispensable part in the field of plant research,plant organs are indispensable in the research of virtual reality.At present,on the processing object of 3D point cloud data,the research on point cloud segmentation of single tree plant organs is not deep enough,the relevant technology is not mature enough,and there is no effective solution for the accurate segmentation of plant organs,especially for the segmentation of petals from a single flower point cloud.Therefore,aiming at the research of single flower point cloud data,this paper proposes a method for feature extraction,segmentation and reconstruction of 3D Flower point cloud data,which provides a reference for plant digital information in the fields of digital city and virtual reality.This paper takes Lily point cloud data as the research object,based on its complex geometric features,proposes to establish a cylindrical coordinate system for it,and carries out research and Implementation on flower point cloud feature extraction,3D point cloud segmentation,3D point cloud reconstruction and other technologies.Based on the traditional K-means clustering point cloud segmentation algorithm,aiming at the uncertainty of K value and the randomness of the initial clustering center,this paper proposes a k-means clustering segmentation algorithm based on flower point cloud feature extraction.Firstly,the central axis of the flower point cloud is extracted,and then the transformation from rectangular coordinate system to cylindrical coordinate system is realized according to the central axis of the point cloud;In the cylindrical coordinate system,the height cross section method and the angle separation method are proposed by using the cylindrical parameter characteristics-height,angle and radius,so as to extract the features of the flower point cloud.Under the flower point cloud feature extraction of height cross-section method and angle separation method,the petal feature of Lily point cloud and the K value in K-means clustering were determined;Then,according to the extracted point cloud features,the cluster set in the height cross-section method is used as the initial cluster center in the K-means segmentation algorithm to segment the petals of the point cloud.Aiming at the phenomenon of over segmentation caused by K-means clustering segmentation,a two partition complement method is proposed,which uses the idea of two partition to segment it,and combines it according to the point cloud label to complete the point cloud.Finally,after analysis and comparison,the greedy projection triangulation algorithm is used to reconstruct the segmented petals.Experiments show that this method can segment petals effectively,and triangulation reconstruction can better preserve the morphological characteristics of flowers with good stability.