| With the continuous advancement of science and technology,3D point cloud reconstruction is increasingly known and applied by people.The research of 3D point cloud reconstruction is widely used in medical,agriculture,entertainment,computer vision,artificial intelligence,unmanned driving,etc.In all fields,3D reconstruction has extremely important research significance and value in many fields.In order to perform three-dimensional point cloud reconstruction,the real data on the surface of the point cloud object must be obtained first,and in the process of three-dimensional scanning,noise point cloud data and data distortion are unavoidably mixed.Therefore,it is the key to the 3D point cloud reconstruction process to propose an effective and efficient filtering algorithm and point cloud reconstruction algorithm.Firstly,a mixed point cloud denoising algorithm is proposed for point cloud with different scales in point cloud data.According to the nature of the noise point,the noise point cloud is divided into a noise point far from the point cloud data and a noise point floating on the surface of the point cloud.A noise removal algorithm is used to denoise the noise points far away from the point cloud data.The data floating on the point cloud surface is subjected to a point cloud bilateral filtering algorithm.Secondly,the point location algorithm based on the area method is proposed to solve the problem that the location point location algorithm is not unique and does not consider the special points in the point cloud reconstruction process.When the 3D point cloud is used for the point location algorithm,it is divided into ordinary points and special points according to the position of the insertion point.When the insertion point is an ordinary point,an improved area point location algorithm is used.When the insertion point is a special point,an improved linear search algorithm is used.Finally,through experiments,compared with the existing point cloud filtering algorithm and point positioning algorithm,the proposed hybrid point cloud denoising algorithm can better perform point cloud denoising while maintaining the properties of the point cloud itself.Good filtering effect and better adaptability,stability and robustness;Point location optimization algorithm based on area method can effectively make up for the problem that the area point location algorithm search path is not unique,and solves the problem of straight line search.The contradiction of triangles improves the efficiency of point-positioning triangles,better constructs the topological relationship among three-dimensional point clouds,and effectively performs point cloud reconstruction. |
