Spatial Point Cloud Data Denoising And Volume Calculation

With the development of science,technology and economy,3D space scanning technology has been developed rapidly,and the accuracy of data obtained by 3D laser scanning instrument is getting higher and higher.The point cloud data of spatial objects often contains gross errors.No matter the specific application of 3D modeling or volume calculation,it needs to be filtered out.Only the clean point cloud and its algorithm exploration based on the volume calculation of spatial objects can be reliable and meaningful.The data obtained by 3D space scanning technology usually contains noise,so it needs to be preprocessed before application.Firstly,an improved k-means denoising algorithm is proposed based on experiments.In this algorithm,firstly,the scattered point cloud is layered,and then the number and band of point cloud are obtained.Then,K-means clustering algorithm is used to cluster and denoise the layered point cloud.The experimental results show that the algorithm has a good denoising effect and can retain the contour shape characteristics of the point cloud model of spatial objects.On this basis,this paper focuses on the volume calculation of irregular objects,adopts the slice method point cloud volume method,and realizes the volume calculation of simple irregular space objects through the improved 360 degree scanning algorithm of plane point cloud ray,two-way nearest point search algorithm of plane point cloud and progressive iterative algorithm based on convex polygon.The research results of this paper are as follows:1.Explore and propose a k-means clustering algorithm for point cloud preprocessing.In this algorithm,firstly,the point cloud is layered,then the K value and initial clustering center of Kmeans algorithm are determined.The K value is the number of layers,and the initial clustering center is the center of gravity of each layer.Finally,the improved k-means clustering algorithm is used to denoise the point cloud.2.Explore and propose solutions to the multi ring effect of point cloud.In this paper,two algorithms are proposed to solve the concentric ring problem and the multi ring problem;the experimental results verify the reliability and accuracy of the two algorithms.3.Slice method is used to calculate the volume of point cloud.The slicing method is to slice the point cloud through the cutting method,then use the boundary search algorithm to determine the point cloud slicing boundary,then calculate the point cloud slicing area,and finally calculate the point cloud volume;the validity and correctness of point cloud volume slicing method are verified by experiments.